cct-cli 0.7.9.0__py3-none-any.whl

calc_terminal/providers/providers.json

@@ -0,0 +1,4075 @@
+{
+  "providers": [
+    {
+      "id": "yi",
+      "name": "01.AI (Yi)",
+      "country": "China",
+      "api_endpoint": "https://api.lingyiwanwu.com/v1",
+      "api_style": "openai",
+      "supports_model_listing": true,
+      "supports_streaming": true,
+      "supports_vision": true,
+      "supports_embeddings": false,
+      "supports_audio": false,
+      "supports_reasoning": false,
+      "openai_compatible": true,
+      "free_models_available": true,
+      "paid_models_available": true,
+      "needs_key": true,
+      "description": "Yi series — 01.AI's flagship model family",
+      "last_updated": "2026-07-31T05:17:14+00:00",
+      "fallback_models": [
+        "yi-large",
+        "yi-medium",
+        "yi-vision",
+        "yi-lightning",
+        "yi-spark",
+        "yi-34b-chat"
+      ],
+      "default_model": "yi-large"
+    },
+    {
+      "id": "abacus",
+      "name": "Abacus AI",
+      "country": "United States",
+      "api_endpoint": "https://api.abacus.ai/v1",
+      "api_style": "openai",
+      "supports_model_listing": true,
+      "supports_streaming": true,
+      "supports_vision": false,
+      "supports_embeddings": false,
+      "supports_audio": false,
+      "supports_reasoning": false,
+      "openai_compatible": true,
+      "free_models_available": true,
+      "paid_models_available": true,
+      "needs_key": true,
+      "description": "Enterprise AI platform",
+      "last_updated": "2026-07-31T05:17:14+00:00",
+      "fallback_models": [
+        "claude-3.5",
+        "gpt-4o"
+      ],
+      "default_model": "gpt-4o"
+    },
+    {
+      "id": "ai21",
+      "name": "AI21 Labs",
+      "country": "Israel",
+      "api_endpoint": "https://api.ai21.com/studio/v1",
+      "api_style": "openai",
+      "supports_model_listing": true,
+      "supports_streaming": true,
+      "supports_vision": false,
+      "supports_embeddings": false,
+      "supports_audio": false,
+      "supports_reasoning": false,
+      "openai_compatible": true,
+      "free_models_available": true,
+      "paid_models_available": true,
+      "needs_key": true,
+      "description": "AI21 Jamba — Hybrid SSM-Transformer",
+      "last_updated": "2026-09-03T00:00:00+00:00",
+      "fallback_models": [
+        "jamba-1.5-large",
+        "jamba-1.5-mini",
+        "jamba-instruct"
+      ],
+      "default_model": "jamba-1.5-large"
+    },
+    {
+      "id": "airtop",
+      "name": "Airtop",
+      "country": "United States",
+      "api_endpoint": "https://api.airtop.ai/v1",
+      "api_style": "openai",
+      "supports_model_listing": true,
+      "supports_streaming": true,
+      "supports_vision": false,
+      "supports_embeddings": false,
+      "supports_audio": false,
+      "supports_reasoning": false,
+      "openai_compatible": true,
+      "free_models_available": true,
+      "paid_models_available": true,
+      "needs_key": true,
+      "description": "AI browser automation",
+      "last_updated": "2026-07-31T05:17:14+00:00",
+      "fallback_models": [
+        "default"
+      ],
+      "default_model": "default"
+    },
+    {
+      "id": "aisera",
+      "name": "Aisera",
+      "country": "United States",
+      "api_endpoint": "https://api.aisera.com/v1",
+      "api_style": "openai",
+      "supports_model_listing": true,
+      "supports_streaming": true,
+      "supports_vision": false,
+      "supports_embeddings": false,
+      "supports_audio": false,
+      "supports_reasoning": false,
+      "openai_compatible": true,
+      "free_models_available": true,
+      "paid_models_available": true,
+      "needs_key": true,
+      "description": "Enterprise AI service desk",
+      "last_updated": "2026-07-31T05:17:14+00:00",
+      "fallback_models": [
+        "aisera-llm"
+      ],
+      "default_model": "aisera-llm"
+    },
+    {
+      "id": "akamai",
+      "name": "Akamai (Cloudlet)",
+      "country": "United States",
+      "api_endpoint": "https://api.cloudlet.ai/v1",
+      "api_style": "openai",
+      "supports_model_listing": true,
+      "supports_streaming": true,
+      "supports_vision": false,
+      "supports_embeddings": false,
+      "supports_audio": false,
+      "supports_reasoning": false,
+      "openai_compatible": true,
+      "free_models_available": true,
+      "paid_models_available": true,
+      "needs_key": true,
+      "description": "Edge AI inference",
+      "last_updated": "2026-07-31T05:17:14+00:00",
+      "fallback_models": [
+        "default"
+      ],
+      "default_model": "default"
+    },
+    {
+      "id": "aleph-alpha",
+      "name": "Aleph Alpha",
+      "country": "Germany",
+      "api_endpoint": "https://api.aleph-alpha.com",
+      "api_style": "openai",
+      "supports_model_listing": true,
+      "supports_streaming": true,
+      "supports_vision": false,
+      "supports_embeddings": true,
+      "supports_audio": false,
+      "supports_reasoning": false,
+      "openai_compatible": true,
+      "free_models_available": true,
+      "paid_models_available": true,
+      "needs_key": true,
+      "description": "Luminous series — European LLM",
+      "last_updated": "2026-07-31T05:17:14+00:00",
+      "fallback_models": [
+        "luminous-base",
+        "luminous-extended",
+        "luminous-supreme"
+      ],
+      "default_model": "luminous-base"
+    },
+    {
+      "id": "alibaba-qwen",
+      "name": "Alibaba Qwen",
+      "country": "China",
+      "api_endpoint": "https://dashscope.aliyuncs.com/compatible-mode/v1",
+      "api_style": "openai",
+      "supports_model_listing": true,
+      "supports_streaming": true,
+      "supports_vision": true,
+      "supports_embeddings": true,
+      "supports_audio": false,
+      "supports_reasoning": true,
+      "openai_compatible": true,
+      "free_models_available": true,
+      "paid_models_available": true,
+      "needs_key": true,
+      "description": "Qwen 2.5 & 3 series — Alibaba's flagship LLMs",
+      "last_updated": "2026-09-03T00:00:00+00:00",
+      "fallback_models": [
+        "qwen-max",
+        "qwen-plus",
+        "qwen-turbo",
+        "qwen2.5-72b-instruct",
+        "qwen2.5-32b-instruct",
+        "qwen2.5-14b-instruct",
+        "qwen-vl-max",
+        "qwen3-235b-a22b"
+      ],
+      "default_model": "qwen-max"
+    },
+    {
+      "id": "ai2",
+      "name": "Allen Institute for AI (OLMo)",
+      "country": "United States",
+      "api_endpoint": "https://api.allenai.org/v1",
+      "api_style": "openai",
+      "supports_model_listing": true,
+      "supports_streaming": true,
+      "supports_vision": true,
+      "supports_embeddings": true,
+      "supports_audio": false,
+      "supports_reasoning": true,
+      "openai_compatible": true,
+      "free_models_available": true,
+      "paid_models_available": false,
+      "needs_key": true,
+      "description": "Open Language Model (OLMo) and Molmo multimodal open-science models",
+      "last_updated": "2026-09-01T00:00:00+00:00",
+      "fallback_models": [
+        "olmo-2-13b-instruct",
+        "olmo-7b-instruct",
+        "molmo-72b-instruct"
+      ],
+      "default_model": "olmo-2-13b-instruct"
+    },
+    {
+      "id": "aws-bedrock",
+      "name": "Amazon Web Services (Bedrock)",
+      "country": "United States",
+      "api_endpoint": "https://bedrock-runtime.us-east-1.amazonaws.com",
+      "api_style": "openai",
+      "supports_model_listing": true,
+      "supports_streaming": true,
+      "supports_vision": false,
+      "supports_embeddings": true,
+      "supports_audio": false,
+      "supports_reasoning": false,
+      "openai_compatible": true,
+      "free_models_available": true,
+      "paid_models_available": true,
+      "needs_key": true,
+      "description": "AWS Bedrock — managed foundation models",
+      "last_updated": "2026-07-31T05:17:14+00:00",
+      "fallback_models": [
+        "claude-3-5-sonnet-20241022",
+        "claude-3-haiku",
+        "llama3-70b",
+        "mistral-large",
+        "anthropic.claude-3-5-sonnet-20241022-v2:0",
+        "anthropic.claude-3-opus-20240229-v1:0",
+        "meta.llama3-1-70b-instruct-v1:0",
+        "amazon.titan-text-express-v1"
+      ],
+      "default_model": "claude-3-5-sonnet-20241022"
+    },
+    {
+      "id": "anthropic",
+      "name": "Anthropic (Claude)",
+      "country": "United States",
+      "api_endpoint": "https://api.anthropic.com/v1",
+      "api_style": "anthropic",
+      "supports_model_listing": true,
+      "supports_streaming": true,
+      "supports_vision": true,
+      "supports_embeddings": false,
+      "supports_audio": false,
+      "supports_reasoning": true,
+      "openai_compatible": false,
+      "free_models_available": true,
+      "paid_models_available": true,
+      "needs_key": true,
+      "description": "Claude 4 Fable & Sonnet — latest AI models",
+      "last_updated": "2026-09-03T00:00:00+00:00",
+      "fallback_models": [
+        "claude-2.1",
+        "claude-3-5-haiku-20241022",
+        "claude-3-5-sonnet-20240620",
+        "claude-3-5-sonnet-20241022",
+        "claude-3-haiku-20240307",
+        "claude-3-opus-20240229",
+        "claude-3-sonnet-20240229",
+        "claude-opus-4",
+        "claude-opus-4-20250514",
+        "claude-sonnet-4",
+        "claude-sonnet-4-20250514",
+        "claude-3-7-sonnet-20250219",
+        "claude-3-5-haiku-latest",
+        "claude-4-fable",
+        "claude-4-fable-5.1"
+      ],
+      "default_model": "claude-4-fable-5.1"
+    },
+    {
+      "id": "anyscale",
+      "name": "Anyscale",
+      "country": "United States",
+      "api_endpoint": "https://api.endpoints.anyscale.com/v1",
+      "api_style": "openai",
+      "supports_model_listing": true,
+      "supports_streaming": true,
+      "supports_vision": false,
+      "supports_embeddings": false,
+      "supports_audio": false,
+      "supports_reasoning": false,
+      "openai_compatible": true,
+      "free_models_available": true,
+      "paid_models_available": true,
+      "needs_key": true,
+      "description": "Ray-based model serving",
+      "last_updated": "2026-07-31T05:17:14+00:00",
+      "fallback_models": [
+        "llama3-70b",
+        "mixtral-8x7b",
+        "meta-llama/Meta-Llama-3.1-70B-Instruct",
+        "mistralai/Mixtral-8x22B-Instruct-v0.1"
+      ],
+      "default_model": "llama3-70b"
+    },
+    {
+      "id": "aphrodite_hosted",
+      "name": "Aphrodite Engine (Local / Remote)",
+      "country": "International",
+      "api_endpoint": "http://localhost:2242/v1",
+      "api_style": "openai",
+      "supports_model_listing": true,
+      "supports_streaming": true,
+      "supports_vision": false,
+      "supports_embeddings": false,
+      "supports_audio": false,
+      "supports_reasoning": true,
+      "openai_compatible": true,
+      "free_models_available": true,
+      "paid_models_available": false,
+      "needs_key": false,
+      "description": "High-throughput LLM serving engine based on vLLM, optimized for creative writing",
+      "last_updated": "2026-09-01T00:00:00+00:00",
+      "fallback_models": [
+        "default",
+        "pygmalion-2-13b",
+        "mythomax-l2-13b"
+      ],
+      "default_model": "default"
+    },
+    {
+      "id": "assemblyai",
+      "name": "AssemblyAI",
+      "country": "United States",
+      "api_endpoint": "https://api.assemblyai.com/v2",
+      "api_style": "openai",
+      "supports_model_listing": true,
+      "supports_streaming": false,
+      "supports_vision": false,
+      "supports_embeddings": false,
+      "supports_audio": true,
+      "supports_reasoning": false,
+      "openai_compatible": false,
+      "free_models_available": true,
+      "paid_models_available": true,
+      "needs_key": true,
+      "description": "Speech-to-text API",
+      "last_updated": "2026-07-31T05:17:14+00:00",
+      "fallback_models": [
+        "default"
+      ],
+      "default_model": "default"
+    },
+    {
+      "id": "azure-openai",
+      "name": "Azure OpenAI",
+      "country": "United States",
+      "api_endpoint": "https://api.openai.azure.com",
+      "api_style": "openai",
+      "supports_model_listing": true,
+      "supports_streaming": true,
+      "supports_vision": true,
+      "supports_embeddings": true,
+      "supports_audio": true,
+      "supports_reasoning": true,
+      "openai_compatible": true,
+      "free_models_available": false,
+      "paid_models_available": true,
+      "needs_key": true,
+      "description": "Azure OpenAI — Enterprise OpenAI models",
+      "last_updated": "2026-09-03T00:00:00+00:00",
+      "fallback_models": [
+        "gpt-4o",
+        "gpt-4o-mini",
+        "gpt-4-turbo",
+        "o3-mini"
+      ],
+      "default_model": "gpt-4o"
+    },
+    {
+      "id": "baichuan",
+      "name": "Baichuan",
+      "country": "China",
+      "api_endpoint": "https://api.baichuan-ai.com/v1",
+      "api_style": "openai",
+      "supports_model_listing": true,
+      "supports_streaming": true,
+      "supports_vision": false,
+      "supports_embeddings": false,
+      "supports_audio": false,
+      "supports_reasoning": false,
+      "openai_compatible": true,
+      "free_models_available": true,
+      "paid_models_available": true,
+      "needs_key": true,
+      "description": "Baichuan 4 — Baichuan's flagship LLM",
+      "last_updated": "2026-09-03T00:00:00+00:00",
+      "fallback_models": [
+        "baichuan4",
+        "baichuan3-turbo",
+        "baichuan2-turbo"
+      ],
+      "default_model": "baichuan4"
+    },
+    {
+      "id": "baidu",
+      "name": "Baidu (ERNIE)",
+      "country": "China",
+      "api_endpoint": "https://aip.baidubce.com/rpc/2.0/ai_custom/v1/wenxinworkshop",
+      "api_style": "openai",
+      "supports_model_listing": true,
+      "supports_streaming": true,
+      "supports_vision": false,
+      "supports_embeddings": false,
+      "supports_audio": false,
+      "supports_reasoning": false,
+      "openai_compatible": true,
+      "free_models_available": true,
+      "paid_models_available": true,
+      "needs_key": true,
+      "description": "ERNIE series — Baidu's flagship LLM",
+      "last_updated": "2026-07-31T05:17:14+00:00",
+      "fallback_models": [
+        "ernie-3.5",
+        "ernie-4.0",
+        "ernie-speed",
+        "ernie-4.0-8k",
+        "ernie-3.5-8k",
+        "ernie-speed-128k",
+        "ernie-lite-8k"
+      ],
+      "default_model": "ernie-3.5"
+    },
+    {
+      "id": "qianfan",
+      "name": "Baidu Qianfan",
+      "country": "China",
+      "api_endpoint": "https://qianfan.baidubce.com/v2",
+      "api_style": "openai",
+      "supports_model_listing": true,
+      "supports_streaming": true,
+      "supports_vision": false,
+      "supports_embeddings": false,
+      "supports_audio": false,
+      "supports_reasoning": false,
+      "openai_compatible": true,
+      "free_models_available": true,
+      "paid_models_available": true,
+      "needs_key": true,
+      "description": "Baidu's enterprise AI platform",
+      "last_updated": "2026-07-31T05:17:14+00:00",
+      "fallback_models": [
+        "ernie-4.0",
+        "ernie-speed"
+      ],
+      "default_model": "ernie-4.0"
+    },
+    {
+      "id": "banana",
+      "name": "Banana",
+      "country": "United States",
+      "api_endpoint": "https://api.banana.dev/v1",
+      "api_style": "openai",
+      "supports_model_listing": true,
+      "supports_streaming": true,
+      "supports_vision": false,
+      "supports_embeddings": false,
+      "supports_audio": false,
+      "supports_reasoning": false,
+      "openai_compatible": true,
+      "free_models_available": true,
+      "paid_models_available": true,
+      "needs_key": true,
+      "description": "Serverless GPU inference",
+      "last_updated": "2026-07-31T05:17:14+00:00",
+      "fallback_models": [
+        "llama3-70b"
+      ],
+      "default_model": "llama3-70b"
+    },
+    {
+      "id": "baseten",
+      "name": "Baseten",
+      "country": "United States",
+      "api_endpoint": "https://bridge.baseten.co/v1",
+      "api_style": "openai",
+      "supports_model_listing": true,
+      "supports_streaming": true,
+      "supports_vision": false,
+      "supports_embeddings": false,
+      "supports_audio": false,
+      "supports_reasoning": false,
+      "openai_compatible": true,
+      "free_models_available": true,
+      "paid_models_available": true,
+      "needs_key": true,
+      "description": "Serverless GPU inference",
+      "last_updated": "2026-07-31T05:17:14+00:00",
+      "fallback_models": [
+        "llama3-70b",
+        "mixtral-8x7b"
+      ],
+      "default_model": "llama3-70b"
+    },
+    {
+      "id": "beam",
+      "name": "Beam",
+      "country": "United States",
+      "api_endpoint": "https://api.beam.cloud/v1",
+      "api_style": "openai",
+      "supports_model_listing": true,
+      "supports_streaming": true,
+      "supports_vision": false,
+      "supports_embeddings": false,
+      "supports_audio": false,
+      "supports_reasoning": false,
+      "openai_compatible": true,
+      "free_models_available": true,
+      "paid_models_available": true,
+      "needs_key": true,
+      "description": "Serverless GPU cloud",
+      "last_updated": "2026-07-31T05:17:14+00:00",
+      "fallback_models": [
+        "llama3-70b"
+      ],
+      "default_model": "llama3-70b"
+    },
+    {
+      "id": "bento_ml",
+      "name": "BentoCloud (BentoML)",
+      "country": "United States",
+      "api_endpoint": "https://cloud.bentoml.com/v1",
+      "api_style": "openai",
+      "supports_model_listing": true,
+      "supports_streaming": true,
+      "supports_vision": true,
+      "supports_embeddings": true,
+      "supports_audio": false,
+      "supports_reasoning": true,
+      "openai_compatible": true,
+      "free_models_available": false,
+      "paid_models_available": true,
+      "needs_key": true,
+      "description": "Production inference platform for fast deployment of open LLMs and custom models",
+      "last_updated": "2026-09-01T00:00:00+00:00",
+      "fallback_models": [
+        "llama3-70b-instruct",
+        "mistral-7b-instruct",
+        "qwen2.5-32b-instruct"
+      ],
+      "default_model": "llama3-70b-instruct"
+    },
+    {
+      "id": "bioage",
+      "name": "BioAge Labs",
+      "country": "United States",
+      "api_endpoint": "https://api.bioage.ai/v1",
+      "api_style": "openai",
+      "supports_model_listing": true,
+      "supports_streaming": true,
+      "supports_vision": false,
+      "supports_embeddings": false,
+      "supports_audio": false,
+      "supports_reasoning": false,
+      "openai_compatible": true,
+      "free_models_available": true,
+      "paid_models_available": true,
+      "needs_key": true,
+      "description": "Biology AI research",
+      "last_updated": "2026-07-31T05:17:14+00:00",
+      "fallback_models": [
+        "default"
+      ],
+      "default_model": "default"
+    },
+    {
+      "id": "black-forest-labs",
+      "name": "Black Forest Labs",
+      "country": "Germany",
+      "api_endpoint": "https://api.black-forest-labs.com/v1",
+      "api_style": "openai",
+      "supports_model_listing": true,
+      "supports_streaming": false,
+      "supports_vision": true,
+      "supports_embeddings": false,
+      "supports_audio": false,
+      "supports_reasoning": false,
+      "openai_compatible": false,
+      "free_models_available": true,
+      "paid_models_available": true,
+      "needs_key": true,
+      "description": "FLUX series — image generation",
+      "last_updated": "2026-07-31T05:17:14+00:00",
+      "fallback_models": [
+        "flux-dev",
+        "flux-pro"
+      ],
+      "default_model": "flux-dev"
+    },
+    {
+      "id": "bytedance-doubao",
+      "name": "ByteDance Doubao",
+      "country": "China",
+      "api_endpoint": "https://ark.cn-beijing.volces.com/api/v3",
+      "api_style": "openai",
+      "supports_model_listing": true,
+      "supports_streaming": true,
+      "supports_vision": true,
+      "supports_embeddings": false,
+      "supports_audio": false,
+      "supports_reasoning": false,
+      "openai_compatible": true,
+      "free_models_available": true,
+      "paid_models_available": true,
+      "needs_key": true,
+      "description": "Doubao Pro — ByteDance's LLM platform",
+      "last_updated": "2026-09-03T00:00:00+00:00",
+      "fallback_models": [
+        "doubao-pro-256k",
+        "doubao-pro-128k",
+        "doubao-lite-128k",
+        "doubao-lite-32k",
+        "doubao-pro-32k"
+      ],
+      "default_model": "doubao-pro-256k"
+    },
+    {
+      "id": "cartesia",
+      "name": "Cartesia Sonic",
+      "country": "United States",
+      "api_endpoint": "https://api.cartesia.ai/v1",
+      "api_style": "openai",
+      "supports_model_listing": true,
+      "supports_streaming": true,
+      "supports_vision": false,
+      "supports_embeddings": false,
+      "supports_audio": true,
+      "supports_reasoning": false,
+      "openai_compatible": false,
+      "free_models_available": false,
+      "paid_models_available": true,
+      "needs_key": true,
+      "description": "Sub-100ms ultra-realistic streaming voice generation",
+      "last_updated": "2026-09-01T00:00:00+00:00",
+      "fallback_models": [
+        "sonic-english",
+        "sonic-multilingual"
+      ],
+      "default_model": "sonic-multilingual"
+    },
+    {
+      "id": "cerebras",
+      "name": "Cerebras",
+      "country": "United States",
+      "api_endpoint": "https://api.cerebras.ai/v1",
+      "api_style": "openai",
+      "supports_model_listing": true,
+      "supports_streaming": true,
+      "supports_vision": false,
+      "supports_embeddings": false,
+      "supports_audio": false,
+      "supports_reasoning": false,
+      "openai_compatible": true,
+      "free_models_available": true,
+      "paid_models_available": true,
+      "needs_key": true,
+      "description": "Cerebras CS-3 — Wafer-scale AI inference",
+      "last_updated": "2026-09-03T00:00:00+00:00",
+      "fallback_models": [
+        "llama3.1-8b",
+        "llama3.1-70b",
+        "llama3.3-70b"
+      ],
+      "default_model": "llama3.3-70b"
+    },
+    {
+      "id": "cerebrium",
+      "name": "Cerebrium",
+      "country": "United States",
+      "api_endpoint": "https://api.cortex.cerebrium.ai/v4",
+      "api_style": "openai",
+      "supports_model_listing": true,
+      "supports_streaming": true,
+      "supports_vision": true,
+      "supports_embeddings": true,
+      "supports_audio": false,
+      "supports_reasoning": true,
+      "openai_compatible": true,
+      "free_models_available": true,
+      "paid_models_available": true,
+      "needs_key": true,
+      "description": "Serverless GPU infrastructure and production inference pipelines for open models",
+      "last_updated": "2026-09-01T00:00:00+00:00",
+      "fallback_models": [
+        "cerebrium-llama-3.1-70b",
+        "cerebrium-mistral-large",
+        "cerebrium-qwen-2.5-72b"
+      ],
+      "default_model": "cerebrium-llama-3.1-70b"
+    },
+    {
+      "id": "chutes",
+      "name": "Chutes.ai",
+      "country": "United States",
+      "api_endpoint": "https://api.chutes.ai/v1",
+      "api_style": "openai",
+      "supports_model_listing": true,
+      "supports_streaming": true,
+      "supports_vision": true,
+      "supports_embeddings": false,
+      "supports_audio": false,
+      "supports_reasoning": true,
+      "openai_compatible": true,
+      "free_models_available": true,
+      "paid_models_available": true,
+      "needs_key": true,
+      "description": "Decentralized serverless compute for open weights and fine-tuned models",
+      "last_updated": "2026-09-01T00:00:00+00:00",
+      "fallback_models": [
+        "deepseek-r1-distill-llama-70b",
+        "llama-3.3-70b-instruct",
+        "qwen-2.5-72b-instruct"
+      ],
+      "default_model": "deepseek-r1-distill-llama-70b"
+    },
+    {
+      "id": "clarifai",
+      "name": "Clarifai",
+      "country": "United States",
+      "api_endpoint": "https://api.clarifai.com/v2",
+      "api_style": "openai",
+      "supports_model_listing": true,
+      "supports_streaming": true,
+      "supports_vision": true,
+      "supports_embeddings": false,
+      "supports_audio": false,
+      "supports_reasoning": false,
+      "openai_compatible": true,
+      "free_models_available": true,
+      "paid_models_available": true,
+      "needs_key": true,
+      "description": "Computer vision & LLM platform",
+      "last_updated": "2026-07-31T05:17:14+00:00",
+      "fallback_models": [
+        "default"
+      ],
+      "default_model": "default"
+    },
+    {
+      "id": "cloudflare_workers_ai",
+      "name": "Cloudflare Workers AI",
+      "country": "United States",
+      "api_endpoint": "https://api.cloudflare.com/client/v4/accounts",
+      "api_style": "openai",
+      "supports_model_listing": true,
+      "supports_streaming": true,
+      "supports_vision": true,
+      "supports_embeddings": true,
+      "supports_audio": true,
+      "supports_reasoning": true,
+      "openai_compatible": true,
+      "free_models_available": true,
+      "paid_models_available": true,
+      "needs_key": true,
+      "description": "Serverless GPU global inference running at Cloudflare edge network",
+      "last_updated": "2026-09-01T00:00:00+00:00",
+      "fallback_models": [
+        "@cf/meta/llama-3.1-70b-instruct",
+        "@cf/deepseek-ai/deepseek-r1-distill-qwen-32b",
+        "@cf/mistral/mistral-7b-instruct-v0.2"
+      ],
+      "default_model": "@cf/meta/llama-3.1-70b-instruct"
+    },
+    {
+      "id": "codeium",
+      "name": "Codeium (Windsurf)",
+      "country": "United States",
+      "api_endpoint": "https://api.codeium.com/v1",
+      "api_style": "openai",
+      "supports_model_listing": true,
+      "supports_streaming": true,
+      "supports_vision": false,
+      "supports_embeddings": false,
+      "supports_audio": false,
+      "supports_reasoning": true,
+      "openai_compatible": true,
+      "free_models_available": true,
+      "paid_models_available": true,
+      "needs_key": true,
+      "description": "Windsurf Cascade and ultra-fast developer reasoning",
+      "last_updated": "2026-09-01T00:00:00+00:00",
+      "fallback_models": [
+        "cascade-base",
+        "codeium-chat",
+        "codeium-flow"
+      ],
+      "default_model": "cascade-base"
+    },
+    {
+      "id": "cohere",
+      "name": "Cohere",
+      "country": "Canada",
+      "api_endpoint": "https://api.cohere.ai/v1",
+      "api_style": "openai",
+      "supports_model_listing": true,
+      "supports_streaming": true,
+      "supports_vision": false,
+      "supports_embeddings": true,
+      "supports_audio": false,
+      "supports_reasoning": false,
+      "openai_compatible": true,
+      "free_models_available": true,
+      "paid_models_available": true,
+      "needs_key": true,
+      "description": "Command-R series & embedding models",
+      "last_updated": "2026-07-31T05:17:14+00:00",
+      "fallback_models": [
+        "command",
+        "command-light",
+        "command-nightly",
+        "command-r",
+        "command-r-plus",
+        "embed-english-v3.0"
+      ],
+      "default_model": "command-r-plus"
+    },
+    {
+      "id": "coreweave",
+      "name": "CoreWeave",
+      "country": "United States",
+      "api_endpoint": "https://api.coreweave.com/v1",
+      "api_style": "openai",
+      "supports_model_listing": true,
+      "supports_streaming": true,
+      "supports_vision": false,
+      "supports_embeddings": false,
+      "supports_audio": false,
+      "supports_reasoning": false,
+      "openai_compatible": true,
+      "free_models_available": true,
+      "paid_models_available": true,
+      "needs_key": true,
+      "description": "GPU cloud for AI",
+      "last_updated": "2026-07-31T05:17:14+00:00",
+      "fallback_models": [
+        "llama3-70b"
+      ],
+      "default_model": "llama3-70b"
+    },
+    {
+      "id": "cortex",
+      "name": "Cortex AI",
+      "country": "United States",
+      "api_endpoint": "https://api.cortex.ai/v1",
+      "api_style": "openai",
+      "supports_model_listing": true,
+      "supports_streaming": true,
+      "supports_vision": false,
+      "supports_embeddings": false,
+      "supports_audio": false,
+      "supports_reasoning": false,
+      "openai_compatible": true,
+      "free_models_available": true,
+      "paid_models_available": true,
+      "needs_key": true,
+      "description": "Enterprise AI platform",
+      "last_updated": "2026-07-31T05:17:14+00:00",
+      "fallback_models": [
+        "default"
+      ],
+      "default_model": "default"
+    },
+    {
+      "id": "cortex_so",
+      "name": "Cortex.so (Local Engine)",
+      "country": "International",
+      "api_endpoint": "http://localhost:39281/v1",
+      "api_style": "openai",
+      "supports_model_listing": true,
+      "supports_streaming": true,
+      "supports_vision": true,
+      "supports_embeddings": true,
+      "supports_audio": false,
+      "supports_reasoning": true,
+      "openai_compatible": true,
+      "free_models_available": true,
+      "paid_models_available": false,
+      "needs_key": false,
+      "description": "Hardware-accelerated local AI runtime supporting multi-engine model orchestration",
+      "last_updated": "2026-09-01T00:00:00+00:00",
+      "fallback_models": [
+        "llama3.1:8b",
+        "mistral:7b",
+        "gemma2:9b"
+      ],
+      "default_model": "llama3.1:8b"
+    },
+    {
+      "id": "cursor",
+      "name": "Cursor (Anysphere)",
+      "country": "United States",
+      "api_endpoint": "https://api.cursor.sh/v1",
+      "api_style": "openai",
+      "supports_model_listing": true,
+      "supports_streaming": true,
+      "supports_vision": true,
+      "supports_embeddings": false,
+      "supports_audio": false,
+      "supports_reasoning": true,
+      "openai_compatible": true,
+      "free_models_available": false,
+      "paid_models_available": true,
+      "needs_key": true,
+      "description": "AI-first code generation and repo-level agent inference",
+      "last_updated": "2026-09-01T00:00:00+00:00",
+      "fallback_models": [
+        "cursor-small",
+        "cursor-fast",
+        "cursor-large"
+      ],
+      "default_model": "cursor-large"
+    },
+    {
+      "id": "databricks",
+      "name": "Databricks",
+      "country": "United States",
+      "api_endpoint": "https://dbc-*.cloud.databricks.com/serving-endpoints",
+      "api_style": "openai",
+      "supports_model_listing": true,
+      "supports_streaming": true,
+      "supports_vision": false,
+      "supports_embeddings": true,
+      "supports_audio": false,
+      "supports_reasoning": false,
+      "openai_compatible": true,
+      "free_models_available": false,
+      "paid_models_available": true,
+      "needs_key": true,
+      "description": "Databricks DBRX — Enterprise AI platform",
+      "last_updated": "2026-09-03T00:00:00+00:00",
+      "fallback_models": [
+        "databricks-dbrx-instruct",
+        "llama-3-70b-chat"
+      ],
+      "default_model": "databricks-dbrx-instruct"
+    },
+    {
+      "id": "deepgram",
+      "name": "Deepgram",
+      "country": "United States",
+      "api_endpoint": "https://api.deepgram.com/v1",
+      "api_style": "openai",
+      "supports_model_listing": true,
+      "supports_streaming": false,
+      "supports_vision": false,
+      "supports_embeddings": false,
+      "supports_audio": true,
+      "supports_reasoning": false,
+      "openai_compatible": false,
+      "free_models_available": true,
+      "paid_models_available": true,
+      "needs_key": true,
+      "description": "Speech recognition & generation",
+      "last_updated": "2026-07-31T05:17:14+00:00",
+      "fallback_models": [
+        "nova-2"
+      ],
+      "default_model": "nova-2"
+    },
+    {
+      "id": "deepinfra",
+      "name": "DeepInfra",
+      "country": "United States",
+      "api_endpoint": "https://api.deepinfra.com/v1/openai",
+      "api_style": "openai",
+      "supports_model_listing": true,
+      "supports_streaming": true,
+      "supports_vision": false,
+      "supports_embeddings": false,
+      "supports_audio": false,
+      "supports_reasoning": false,
+      "openai_compatible": true,
+      "free_models_available": true,
+      "paid_models_available": true,
+      "needs_key": true,
+      "description": "DeepInfra — Serverless AI inference",
+      "last_updated": "2026-09-03T00:00:00+00:00",
+      "fallback_models": [
+        "meta-llama/Meta-Llama-3.1-405B-Instruct",
+        "meta-llama/Meta-Llama-3.1-70B-Instruct",
+        "mistralai/Mixtral-8x22B-Instruct-v0.1",
+        "Qwen/Qwen2.5-72B-Instruct"
+      ],
+      "default_model": "meta-llama/Meta-Llama-3.1-70B-Instruct"
+    },
+    {
+      "id": "deepseek",
+      "name": "DeepSeek",
+      "country": "China",
+      "api_endpoint": "https://api.deepseek.com/v1",
+      "api_style": "openai",
+      "supports_model_listing": true,
+      "supports_streaming": true,
+      "supports_vision": false,
+      "supports_embeddings": false,
+      "supports_audio": false,
+      "supports_reasoning": true,
+      "openai_compatible": true,
+      "free_models_available": true,
+      "paid_models_available": true,
+      "needs_key": true,
+      "description": "DeepSeek-V3 & DeepSeek-R1 — strong reasoning",
+      "last_updated": "2026-09-03T00:00:00+00:00",
+      "fallback_models": [
+        "deepseek-chat",
+        "deepseek-coder",
+        "deepseek-reasoner",
+        "deepseek-v3",
+        "deepseek-r1"
+      ],
+      "default_model": "deepseek-chat"
+    },
+    {
+      "id": "deployai",
+      "name": "Deploy AI",
+      "country": "United States",
+      "api_endpoint": "https://api.deploy.ai/v1",
+      "api_style": "openai",
+      "supports_model_listing": true,
+      "supports_streaming": true,
+      "supports_vision": false,
+      "supports_embeddings": false,
+      "supports_audio": false,
+      "supports_reasoning": false,
+      "openai_compatible": true,
+      "free_models_available": true,
+      "paid_models_available": true,
+      "needs_key": true,
+      "description": "Model deployment platform",
+      "last_updated": "2026-07-31T05:17:14+00:00",
+      "fallback_models": [
+        "default"
+      ],
+      "default_model": "default"
+    },
+    {
+      "id": "elastic",
+      "name": "Elastic (ELSER)",
+      "country": "United States",
+      "api_endpoint": "https://api.elastic-cloud.com/v1",
+      "api_style": "openai",
+      "supports_model_listing": true,
+      "supports_streaming": false,
+      "supports_vision": false,
+      "supports_embeddings": true,
+      "supports_audio": false,
+      "supports_reasoning": false,
+      "openai_compatible": true,
+      "free_models_available": true,
+      "paid_models_available": true,
+      "needs_key": true,
+      "description": "Elasticsearch AI search",
+      "last_updated": "2026-07-31T05:17:14+00:00",
+      "fallback_models": [
+        "elser-v2"
+      ],
+      "default_model": "elser-v2"
+    },
+    {
+      "id": "eleutherai",
+      "name": "EleutherAI",
+      "country": "International",
+      "api_endpoint": "https://api.eleuther.ai/v1",
+      "api_style": "openai",
+      "supports_model_listing": true,
+      "supports_streaming": true,
+      "supports_vision": false,
+      "supports_embeddings": false,
+      "supports_audio": false,
+      "supports_reasoning": false,
+      "openai_compatible": true,
+      "free_models_available": true,
+      "paid_models_available": false,
+      "needs_key": false,
+      "description": "Open-source AI research",
+      "last_updated": "2026-07-31T05:17:14+00:00",
+      "fallback_models": [
+        "pythia-12b"
+      ],
+      "default_model": "pythia-12b"
+    },
+    {
+      "id": "elevenlabs",
+      "name": "ElevenLabs AI",
+      "country": "United States",
+      "api_endpoint": "https://api.elevenlabs.io/v1",
+      "api_style": "openai",
+      "supports_model_listing": true,
+      "supports_streaming": true,
+      "supports_vision": false,
+      "supports_embeddings": false,
+      "supports_audio": true,
+      "supports_reasoning": false,
+      "openai_compatible": false,
+      "free_models_available": true,
+      "paid_models_available": true,
+      "needs_key": true,
+      "description": "Industry leader in voice synthesis, voice cloning, and audio AI",
+      "last_updated": "2026-09-01T00:00:00+00:00",
+      "fallback_models": [
+        "eleven_multilingual_v2",
+        "eleven_turbo_v2_5",
+        "eleven_flash_v2"
+      ],
+      "default_model": "eleven_multilingual_v2"
+    },
+    {
+      "id": "exllamav2_hosted",
+      "name": "ExLlamaV2 Server (Local)",
+      "country": "International",
+      "api_endpoint": "http://localhost:5000/v1",
+      "api_style": "openai",
+      "supports_model_listing": true,
+      "supports_streaming": true,
+      "supports_vision": false,
+      "supports_embeddings": false,
+      "supports_audio": false,
+      "supports_reasoning": true,
+      "openai_compatible": true,
+      "free_models_available": true,
+      "paid_models_available": false,
+      "needs_key": false,
+      "description": "Ultra-fast inference library designed from scratch for local modern GPUs",
+      "last_updated": "2026-09-01T00:00:00+00:00",
+      "fallback_models": [
+        "exllamav2",
+        "llama-3.1-70b-exl2"
+      ],
+      "default_model": "exllamav2"
+    },
+    {
+      "id": "fal_ai",
+      "name": "fal.ai",
+      "country": "United States",
+      "api_endpoint": "https://queue.fal.run",
+      "api_style": "openai",
+      "supports_model_listing": true,
+      "supports_streaming": true,
+      "supports_vision": true,
+      "supports_embeddings": false,
+      "supports_audio": true,
+      "supports_reasoning": false,
+      "openai_compatible": true,
+      "free_models_available": true,
+      "paid_models_available": true,
+      "needs_key": true,
+      "description": "Ultra-low latency serverless media, vision, and multimodal inference platform",
+      "last_updated": "2026-09-01T00:00:00+00:00",
+      "fallback_models": [
+        "flux-pro-1.1",
+        "fast-sdxl",
+        "whisper-large-v3"
+      ],
+      "default_model": "flux-pro-1.1"
+    },
+    {
+      "id": "featherless",
+      "name": "Featherless AI",
+      "country": "United States",
+      "api_endpoint": "https://api.featherless.ai/v1",
+      "api_style": "openai",
+      "supports_model_listing": true,
+      "supports_streaming": true,
+      "supports_vision": false,
+      "supports_embeddings": false,
+      "supports_audio": false,
+      "supports_reasoning": false,
+      "openai_compatible": true,
+      "free_models_available": true,
+      "paid_models_available": true,
+      "needs_key": true,
+      "description": "Simple model API",
+      "last_updated": "2026-07-31T05:17:14+00:00",
+      "fallback_models": [
+        "llama3-70b"
+      ],
+      "default_model": "llama3-70b"
+    },
+    {
+      "id": "fireworks",
+      "name": "Fireworks AI",
+      "country": "United States",
+      "api_endpoint": "https://api.fireworks.ai/inference/v1",
+      "api_style": "openai",
+      "supports_model_listing": true,
+      "supports_streaming": true,
+      "supports_vision": true,
+      "supports_embeddings": true,
+      "supports_audio": false,
+      "supports_reasoning": false,
+      "openai_compatible": true,
+      "free_models_available": true,
+      "paid_models_available": true,
+      "needs_key": true,
+      "description": "Fireworks AI — Fast inference platform",
+      "last_updated": "2026-09-03T00:00:00+00:00",
+      "fallback_models": [
+        "accounts/fireworks/models/llama-v3p3-70b-instruct",
+        "accounts/fireworks/models/mixtral-8x22b-instruct",
+        "accounts/fireworks/models/qwen2-72b-instruct",
+        "accounts/fireworks/models/deepseek-v3"
+      ],
+      "default_model": "accounts/fireworks/models/llama-v3p3-70b-instruct"
+    },
+    {
+      "id": "forefront",
+      "name": "Forefront AI",
+      "country": "United States",
+      "api_endpoint": "https://api.forefront.ai/v1",
+      "api_style": "openai",
+      "supports_model_listing": true,
+      "supports_streaming": true,
+      "supports_vision": false,
+      "supports_embeddings": false,
+      "supports_audio": false,
+      "supports_reasoning": false,
+      "openai_compatible": true,
+      "free_models_available": true,
+      "paid_models_available": true,
+      "needs_key": true,
+      "description": "Multi-provider API gateway",
+      "last_updated": "2026-07-31T05:17:14+00:00",
+      "fallback_models": [
+        "claude-3",
+        "gpt-4o"
+      ],
+      "default_model": "gpt-4o"
+    },
+    {
+      "id": "friendli",
+      "name": "FriendliAI",
+      "country": "South Korea",
+      "api_endpoint": "https://inference.friendli.ai/v1",
+      "api_style": "openai",
+      "supports_model_listing": true,
+      "supports_streaming": true,
+      "supports_vision": false,
+      "supports_embeddings": false,
+      "supports_audio": false,
+      "supports_reasoning": false,
+      "openai_compatible": true,
+      "free_models_available": true,
+      "paid_models_available": true,
+      "needs_key": true,
+      "description": "FriendliAI — Fast inference platform",
+      "last_updated": "2026-09-03T00:00:00+00:00",
+      "fallback_models": [
+        "meta-llama-3.1-8b-instruct",
+        "meta-llama-3.1-70b-instruct",
+        "mixtral-8x7b-instruct-v0.1"
+      ],
+      "default_model": "meta-llama-3.1-70b-instruct"
+    },
+    {
+      "id": "gcore",
+      "name": "Gcore",
+      "country": "Luxembourg",
+      "api_endpoint": "https://api.gcore.com/v1",
+      "api_style": "openai",
+      "supports_model_listing": true,
+      "supports_streaming": true,
+      "supports_vision": false,
+      "supports_embeddings": false,
+      "supports_audio": false,
+      "supports_reasoning": false,
+      "openai_compatible": true,
+      "free_models_available": true,
+      "paid_models_available": true,
+      "needs_key": true,
+      "description": "Edge & cloud AI inference",
+      "last_updated": "2026-07-31T05:17:14+00:00",
+      "fallback_models": [
+        "llama3-70b"
+      ],
+      "default_model": "llama3-70b"
+    },
+    {
+      "id": "github-copilot",
+      "name": "GitHub Copilot API",
+      "country": "United States",
+      "api_endpoint": "https://api.githubcopilot.com",
+      "api_style": "openai",
+      "supports_model_listing": true,
+      "supports_streaming": true,
+      "supports_vision": true,
+      "supports_embeddings": true,
+      "supports_audio": false,
+      "supports_reasoning": true,
+      "openai_compatible": true,
+      "free_models_available": false,
+      "paid_models_available": true,
+      "needs_key": true,
+      "description": "GitHub Copilot developer models — GPT-4o, Claude 3.5, Gemini 1.5",
+      "last_updated": "2026-09-01T00:00:00+00:00",
+      "fallback_models": [
+        "gpt-4o",
+        "claude-3.5-sonnet",
+        "o1-preview"
+      ],
+      "default_model": "gpt-4o"
+    },
+    {
+      "id": "gemini",
+      "name": "Google Gemini",
+      "country": "United States",
+      "api_endpoint": "https://generativelanguage.googleapis.com/v1beta",
+      "api_style": "gemini",
+      "supports_model_listing": true,
+      "supports_streaming": true,
+      "supports_vision": true,
+      "supports_embeddings": true,
+      "supports_audio": false,
+      "supports_reasoning": true,
+      "openai_compatible": false,
+      "free_models_available": true,
+      "paid_models_available": true,
+      "needs_key": true,
+      "description": "Gemini 2.0 & 2.5 — Google DeepMind",
+      "last_updated": "2026-09-03T00:00:00+00:00",
+      "fallback_models": [
+        "gemini-1.5-flash",
+        "gemini-1.5-flash-8b",
+        "gemini-1.5-flash-latest",
+        "gemini-1.5-pro",
+        "gemini-1.5-pro-latest",
+        "gemini-2.0-flash",
+        "gemini-2.0-flash-exp",
+        "gemini-2.5-flash",
+        "gemini-2.5-flash-preview-04-17",
+        "gemini-2.5-pro",
+        "gemini-2.5-pro-exp-03-25",
+        "gemini-2.5-pro-preview-05-06",
+        "gemma-2-27b-it",
+        "gemma-2-9b-it"
+      ],
+      "default_model": "gemini-2.5-flash"
+    },
+    {
+      "id": "palm",
+      "name": "Google PaLM API",
+      "country": "United States",
+      "api_endpoint": "https://generativelanguage.googleapis.com/v1beta2",
+      "api_style": "gemini",
+      "supports_model_listing": true,
+      "supports_streaming": true,
+      "supports_vision": false,
+      "supports_embeddings": false,
+      "supports_audio": false,
+      "supports_reasoning": false,
+      "openai_compatible": false,
+      "free_models_available": true,
+      "paid_models_available": true,
+      "needs_key": true,
+      "description": "PaLM 2 — Google's previous-gen LLM",
+      "last_updated": "2026-07-31T05:17:14+00:00",
+      "fallback_models": [
+        "chat-bison-001",
+        "text-bison-001"
+      ],
+      "default_model": "chat-bison-001"
+    },
+    {
+      "id": "google-vertex",
+      "name": "Google Vertex AI",
+      "country": "United States",
+      "api_endpoint": "https://us-central1-aiplatform.googleapis.com/v1",
+      "api_style": "gemini",
+      "supports_model_listing": true,
+      "supports_streaming": true,
+      "supports_vision": true,
+      "supports_embeddings": true,
+      "supports_audio": false,
+      "supports_reasoning": true,
+      "openai_compatible": false,
+      "free_models_available": false,
+      "paid_models_available": true,
+      "needs_key": true,
+      "description": "Google Vertex AI — Enterprise Gemini",
+      "last_updated": "2026-09-03T00:00:00+00:00",
+      "fallback_models": [
+        "gemini-2.5-pro",
+        "gemini-2.5-flash",
+        "gemini-2.0-flash",
+        "text-embedding-004",
+        "claude-3-5-sonnet",
+        "gemini-1.5-pro",
+        "llama3-70b"
+      ],
+      "default_model": "gemini-2.5-pro"
+    },
+    {
+      "id": "gpt4all",
+      "name": "GPT4All",
+      "country": "International",
+      "api_endpoint": "http://localhost:4891",
+      "api_style": "ollama",
+      "supports_model_listing": true,
+      "supports_streaming": true,
+      "supports_vision": false,
+      "supports_embeddings": true,
+      "supports_audio": false,
+      "supports_reasoning": false,
+      "openai_compatible": true,
+      "free_models_available": true,
+      "paid_models_available": false,
+      "needs_key": false,
+      "description": "GPT4All — Local AI assistant",
+      "last_updated": "2026-09-03T00:00:00+00:00",
+      "fallback_models": [
+        "gpt4all-falcon",
+        "gpt4all-llama3-8b",
+        "gpt4all-mistral-7b"
+      ],
+      "default_model": "gpt4all-llama3-8b"
+    },
+    {
+      "id": "gradient",
+      "name": "Gradient AI",
+      "country": "Canada",
+      "api_endpoint": "https://api.gradient.ai/v1",
+      "api_style": "openai",
+      "supports_model_listing": true,
+      "supports_streaming": true,
+      "supports_vision": false,
+      "supports_embeddings": false,
+      "supports_audio": false,
+      "supports_reasoning": false,
+      "openai_compatible": true,
+      "free_models_available": true,
+      "paid_models_available": true,
+      "needs_key": true,
+      "description": "Serverless fine-tuning & inference",
+      "last_updated": "2026-07-31T05:17:14+00:00",
+      "fallback_models": [
+        "llama3-70b"
+      ],
+      "default_model": "llama3-70b"
+    },
+    {
+      "id": "groq",
+      "name": "Groq",
+      "country": "United States",
+      "api_endpoint": "https://api.groq.com/openai/v1",
+      "api_style": "openai",
+      "supports_model_listing": true,
+      "supports_streaming": true,
+      "supports_vision": false,
+      "supports_embeddings": false,
+      "supports_audio": false,
+      "supports_reasoning": false,
+      "openai_compatible": true,
+      "free_models_available": true,
+      "paid_models_available": true,
+      "needs_key": true,
+      "description": "LPU inference — exceptionally fast",
+      "last_updated": "2026-07-31T05:17:14+00:00",
+      "fallback_models": [
+        "deepseek-r1-distill-llama-70b",
+        "gemma2-9b-it",
+        "llama-3.1-70b-versatile",
+        "llama-3.1-8b-instant",
+        "llama-3.3-70b-versatile",
+        "llama-4-scout-17b-16e-instruct",
+        "llama3-70b-8192",
+        "llama3.1-70b-versatile",
+        "mixtral-8x7b-32768"
+      ],
+      "default_model": "llama-3.3-70b-versatile"
+    },
+    {
+      "id": "h2o",
+      "name": "H2O.ai",
+      "country": "United States",
+      "api_endpoint": "https://api.h2o.ai/v1",
+      "api_style": "openai",
+      "supports_model_listing": true,
+      "supports_streaming": true,
+      "supports_vision": false,
+      "supports_embeddings": false,
+      "supports_audio": false,
+      "supports_reasoning": false,
+      "openai_compatible": true,
+      "free_models_available": true,
+      "paid_models_available": true,
+      "needs_key": true,
+      "description": "Enterprise AI & h2oGPT",
+      "last_updated": "2026-07-31T05:17:14+00:00",
+      "fallback_models": [
+        "h2ogpt-gm"
+      ],
+      "default_model": "h2ogpt-gm"
+    },
+    {
+      "id": "hotpot",
+      "name": "Hotpot AI",
+      "country": "United States",
+      "api_endpoint": "https://api.hotpot.ai/v1",
+      "api_style": "openai",
+      "supports_model_listing": true,
+      "supports_streaming": true,
+      "supports_vision": true,
+      "supports_embeddings": false,
+      "supports_audio": false,
+      "supports_reasoning": false,
+      "openai_compatible": true,
+      "free_models_available": true,
+      "paid_models_available": true,
+      "needs_key": true,
+      "description": "AI image editing & generation",
+      "last_updated": "2026-07-31T05:17:14+00:00",
+      "fallback_models": [
+        "default"
+      ],
+      "default_model": "default"
+    },
+    {
+      "id": "huggingface",
+      "name": "Hugging Face",
+      "country": "United States",
+      "api_endpoint": "https://api-inference.huggingface.co",
+      "api_style": "openai",
+      "supports_model_listing": true,
+      "supports_streaming": true,
+      "supports_vision": true,
+      "supports_embeddings": true,
+      "supports_audio": false,
+      "supports_reasoning": false,
+      "openai_compatible": true,
+      "free_models_available": true,
+      "paid_models_available": true,
+      "needs_key": true,
+      "description": "Hugging Face Inference — Open-source model hub",
+      "last_updated": "2026-09-03T00:00:00+00:00",
+      "fallback_models": [
+        "meta-llama/Llama-3.1-405B-Instruct",
+        "meta-llama/Llama-3.1-70B-Instruct",
+        "mistralai/Mixtral-8x22B-Instruct-v0.1",
+        "Qwen/Qwen2.5-72B-Instruct"
+      ],
+      "default_model": "meta-llama/Llama-3.1-70B-Instruct"
+    },
+    {
+      "id": "tgi_hosted",
+      "name": "Hugging Face TGI (Text Generation Inference)",
+      "country": "United States",
+      "api_endpoint": "http://localhost:8080/v1",
+      "api_style": "openai",
+      "supports_model_listing": true,
+      "supports_streaming": true,
+      "supports_vision": true,
+      "supports_embeddings": false,
+      "supports_audio": false,
+      "supports_reasoning": true,
+      "openai_compatible": true,
+      "free_models_available": true,
+      "paid_models_available": false,
+      "needs_key": false,
+      "description": "Purpose-built solution for deploying and serving large language models",
+      "last_updated": "2026-09-01T00:00:00+00:00",
+      "fallback_models": [
+        "tgi",
+        "mistralai/Mistral-7B-Instruct-v0.3",
+        "meta-llama/Meta-Llama-3-8B-Instruct"
+      ],
+      "default_model": "tgi"
+    },
+    {
+      "id": "hyperbolic",
+      "name": "Hyperbolic",
+      "country": "United States",
+      "api_endpoint": "https://api.hyperbolic.xyz/v1",
+      "api_style": "openai",
+      "supports_model_listing": true,
+      "supports_streaming": true,
+      "supports_vision": false,
+      "supports_embeddings": false,
+      "supports_audio": false,
+      "supports_reasoning": false,
+      "openai_compatible": true,
+      "free_models_available": true,
+      "paid_models_available": true,
+      "needs_key": true,
+      "description": "Open model inference",
+      "last_updated": "2026-07-31T05:17:14+00:00",
+      "fallback_models": [
+        "llama3-70b",
+        "qwen2-72b",
+        "meta-llama/Llama-3.3-70B-Instruct",
+        "deepseek-ai/DeepSeek-V3"
+      ],
+      "default_model": "llama3-70b"
+    },
+    {
+      "id": "ibm-watsonx",
+      "name": "IBM watsonx",
+      "country": "United States",
+      "api_endpoint": "https://us-south.ml.cloud.ibm.com/ml/v1",
+      "api_style": "openai",
+      "supports_model_listing": true,
+      "supports_streaming": true,
+      "supports_vision": false,
+      "supports_embeddings": true,
+      "supports_audio": false,
+      "supports_reasoning": false,
+      "openai_compatible": true,
+      "free_models_available": false,
+      "paid_models_available": true,
+      "needs_key": true,
+      "description": "IBM watsonx — Enterprise AI platform",
+      "last_updated": "2026-09-03T00:00:00+00:00",
+      "fallback_models": [
+        "ibm/granite-3-8b-instruct",
+        "ibm/granite-3-2b-instruct",
+        "meta-llama/llama-3-1-70b-instruct"
+      ],
+      "default_model": "ibm/granite-3-8b-instruct"
+    },
+    {
+      "id": "watson",
+      "name": "IBM Watsonx.ai",
+      "country": "United States",
+      "api_endpoint": "https://api.watsonx.ai/v1",
+      "api_style": "openai",
+      "supports_model_listing": true,
+      "supports_streaming": true,
+      "supports_vision": false,
+      "supports_embeddings": false,
+      "supports_audio": false,
+      "supports_reasoning": false,
+      "openai_compatible": true,
+      "free_models_available": true,
+      "paid_models_available": true,
+      "needs_key": true,
+      "description": "IBM enterprise AI platform",
+      "last_updated": "2026-07-31T05:17:14+00:00",
+      "fallback_models": [
+        "granite-13b",
+        "llama3-70b"
+      ],
+      "default_model": "granite-13b"
+    },
+    {
+      "id": "idiomatic",
+      "name": "Idiomatic AI",
+      "country": "Canada",
+      "api_endpoint": "https://api.idiomatic.ai/v1",
+      "api_style": "openai",
+      "supports_model_listing": true,
+      "supports_streaming": true,
+      "supports_vision": false,
+      "supports_embeddings": false,
+      "supports_audio": false,
+      "supports_reasoning": false,
+      "openai_compatible": true,
+      "free_models_available": true,
+      "paid_models_available": true,
+      "needs_key": true,
+      "description": "NLP for customer experience",
+      "last_updated": "2026-07-31T05:17:14+00:00",
+      "fallback_models": [
+        "default"
+      ],
+      "default_model": "default"
+    },
+    {
+      "id": "xunfei-spark",
+      "name": "iFlytek Spark",
+      "country": "China",
+      "api_endpoint": "https://spark-api-open.xf-yun.com/v1",
+      "api_style": "openai",
+      "supports_model_listing": true,
+      "supports_streaming": true,
+      "supports_vision": true,
+      "supports_embeddings": false,
+      "supports_audio": false,
+      "supports_reasoning": false,
+      "openai_compatible": true,
+      "free_models_available": true,
+      "paid_models_available": true,
+      "needs_key": true,
+      "description": "Spark 4.0 — iFlytek's multimodal LLM",
+      "last_updated": "2026-09-03T00:00:00+00:00",
+      "fallback_models": [
+        "spark4.0-ultra",
+        "spark-max",
+        "spark-pro",
+        "spark-lite"
+      ],
+      "default_model": "spark4.0-ultra"
+    },
+    {
+      "id": "infermatic",
+      "name": "Infermatic.ai",
+      "country": "United States",
+      "api_endpoint": "https://api.infermatic.ai/v1",
+      "api_style": "openai",
+      "supports_model_listing": true,
+      "supports_streaming": true,
+      "supports_vision": true,
+      "supports_embeddings": false,
+      "supports_audio": false,
+      "supports_reasoning": true,
+      "openai_compatible": true,
+      "free_models_available": false,
+      "paid_models_available": true,
+      "needs_key": true,
+      "description": "Platform specialized in open-weights models, storytelling, roleplay, and uncensored LLMs",
+      "last_updated": "2026-09-01T00:00:00+00:00",
+      "fallback_models": [
+        "magnum-v4-72b",
+        "command-r-plus",
+        "llama-3.1-nemotron-70b"
+      ],
+      "default_model": "magnum-v4-72b"
+    },
+    {
+      "id": "inflection",
+      "name": "Inflection AI",
+      "country": "United States",
+      "api_endpoint": "https://api.inflection.ai/v1",
+      "api_style": "openai",
+      "supports_model_listing": true,
+      "supports_streaming": true,
+      "supports_vision": false,
+      "supports_embeddings": false,
+      "supports_audio": false,
+      "supports_reasoning": false,
+      "openai_compatible": true,
+      "free_models_available": true,
+      "paid_models_available": true,
+      "needs_key": true,
+      "description": "Pi — personal AI",
+      "last_updated": "2026-07-31T05:17:14+00:00",
+      "fallback_models": [
+        "inflection-2.5"
+      ],
+      "default_model": "inflection-2.5"
+    },
+    {
+      "id": "internlm",
+      "name": "InternLM (Shanghai AI Lab)",
+      "country": "China",
+      "api_endpoint": "https://internlm-chat.intern-ai.org.cn/api/v1",
+      "api_style": "openai",
+      "supports_model_listing": true,
+      "supports_streaming": true,
+      "supports_vision": true,
+      "supports_embeddings": false,
+      "supports_audio": false,
+      "supports_reasoning": true,
+      "openai_compatible": true,
+      "free_models_available": true,
+      "paid_models_available": true,
+      "needs_key": true,
+      "description": "Open-source bilingual foundation models with strong reasoning and coding capabilities",
+      "last_updated": "2026-09-01T00:00:00+00:00",
+      "fallback_models": [
+        "internlm2_5-20b-chat",
+        "internlm2_5-7b-chat",
+        "internlm-xcomposer2d5-7b"
+      ],
+      "default_model": "internlm2_5-20b-chat"
+    },
+    {
+      "id": "jan_ai",
+      "name": "Jan.ai (Local Server)",
+      "country": "Singapore",
+      "api_endpoint": "http://localhost:1337/v1",
+      "api_style": "openai",
+      "supports_model_listing": true,
+      "supports_streaming": true,
+      "supports_vision": true,
+      "supports_embeddings": false,
+      "supports_audio": false,
+      "supports_reasoning": true,
+      "openai_compatible": true,
+      "free_models_available": true,
+      "paid_models_available": false,
+      "needs_key": false,
+      "description": "Open source, local-first alternative to ChatGPT running 100% offline on your device",
+      "last_updated": "2026-09-01T00:00:00+00:00",
+      "fallback_models": [
+        "llama3.1-8b-instruct",
+        "deepseek-coder-6.7b",
+        "mistral-ins-7b-q4"
+      ],
+      "default_model": "llama3.1-8b-instruct"
+    },
+    {
+      "id": "jasper",
+      "name": "Jasper AI",
+      "country": "United States",
+      "api_endpoint": "https://api.jasper.ai/v1",
+      "api_style": "openai",
+      "supports_model_listing": true,
+      "supports_streaming": true,
+      "supports_vision": false,
+      "supports_embeddings": false,
+      "supports_audio": false,
+      "supports_reasoning": false,
+      "openai_compatible": true,
+      "free_models_available": true,
+      "paid_models_available": true,
+      "needs_key": true,
+      "description": "AI content platform",
+      "last_updated": "2026-07-31T05:17:14+00:00",
+      "fallback_models": [
+        "default"
+      ],
+      "default_model": "default"
+    },
+    {
+      "id": "jina",
+      "name": "Jina AI",
+      "country": "Germany",
+      "api_endpoint": "https://api.jina.ai/v1",
+      "api_style": "openai",
+      "supports_model_listing": true,
+      "supports_streaming": false,
+      "supports_vision": false,
+      "supports_embeddings": true,
+      "supports_audio": false,
+      "supports_reasoning": false,
+      "openai_compatible": true,
+      "free_models_available": true,
+      "paid_models_available": true,
+      "needs_key": true,
+      "description": "Embeddings & neural search",
+      "last_updated": "2026-07-31T05:17:14+00:00",
+      "fallback_models": [
+        "jina-embeddings-v3",
+        "reader-lm"
+      ],
+      "default_model": "jina-embeddings-v3"
+    },
+    {
+      "id": "kakaocorp",
+      "name": "Kakao Corp (HyperCLOVA)",
+      "country": "South Korea",
+      "api_endpoint": "https://api.kakaocloud.com/v1",
+      "api_style": "openai",
+      "supports_model_listing": true,
+      "supports_streaming": true,
+      "supports_vision": false,
+      "supports_embeddings": false,
+      "supports_audio": false,
+      "supports_reasoning": false,
+      "openai_compatible": true,
+      "free_models_available": true,
+      "paid_models_available": true,
+      "needs_key": true,
+      "description": "HyperCLOVA X — Kakao's Korean-optimized LLM",
+      "last_updated": "2026-09-03T00:00:00+00:00",
+      "fallback_models": [
+        "hyperclova-x-seed-llama-3.1-70b",
+        "hyperclova-x-seed-llama-3.1-8b"
+      ],
+      "default_model": "hyperclova-x-seed-llama-3.1-70b"
+    },
+    {
+      "id": "koala",
+      "name": "Koala AI",
+      "country": "United States",
+      "api_endpoint": "https://api.koala.sh/v1",
+      "api_style": "openai",
+      "supports_model_listing": true,
+      "supports_streaming": true,
+      "supports_vision": false,
+      "supports_embeddings": false,
+      "supports_audio": false,
+      "supports_reasoning": false,
+      "openai_compatible": true,
+      "free_models_available": true,
+      "paid_models_available": true,
+      "needs_key": true,
+      "description": "Fine-tuned chat model",
+      "last_updated": "2026-07-31T05:17:14+00:00",
+      "fallback_models": [
+        "koala-13b"
+      ],
+      "default_model": "koala-13b"
+    },
+    {
+      "id": "koboldai",
+      "name": "KoboldAI",
+      "country": "International",
+      "api_endpoint": "https://lite.koboldai.net/v1",
+      "api_style": "openai",
+      "supports_model_listing": true,
+      "supports_streaming": true,
+      "supports_vision": false,
+      "supports_embeddings": false,
+      "supports_audio": false,
+      "supports_reasoning": false,
+      "openai_compatible": true,
+      "free_models_available": true,
+      "paid_models_available": false,
+      "needs_key": false,
+      "description": "Community-run open model inference",
+      "last_updated": "2026-07-31T05:17:14+00:00",
+      "fallback_models": [
+        "mythomax-l2-13b",
+        "pygmalion-6b"
+      ],
+      "default_model": "mythomax-l2-13b"
+    },
+    {
+      "id": "koboldcpp",
+      "name": "KoboldCpp",
+      "country": "International",
+      "api_endpoint": "http://localhost:5001/v1",
+      "api_style": "openai",
+      "supports_model_listing": true,
+      "supports_streaming": true,
+      "supports_vision": false,
+      "supports_embeddings": false,
+      "supports_audio": false,
+      "supports_reasoning": false,
+      "openai_compatible": true,
+      "free_models_available": true,
+      "paid_models_available": false,
+      "needs_key": false,
+      "description": "KoboldCpp — Local LLM runner",
+      "last_updated": "2026-09-03T00:00:00+00:00",
+      "fallback_models": [
+        "default",
+        "kobold-model",
+        "fimbulvetr-11b-v2"
+      ],
+      "default_model": "default"
+    },
+    {
+      "id": "kyutai",
+      "name": "Kyutai (Moshi)",
+      "country": "France",
+      "api_endpoint": "https://api.kyutai.org/v1",
+      "api_style": "openai",
+      "supports_model_listing": true,
+      "supports_streaming": true,
+      "supports_vision": false,
+      "supports_embeddings": false,
+      "supports_audio": true,
+      "supports_reasoning": true,
+      "openai_compatible": true,
+      "free_models_available": true,
+      "paid_models_available": false,
+      "needs_key": false,
+      "description": "Full-duplex real-time multimodal voice and conversational AI foundation lab",
+      "last_updated": "2026-09-01T00:00:00+00:00",
+      "fallback_models": [
+        "moshi-v1",
+        "mimi-audio-codec",
+        "helium-1"
+      ],
+      "default_model": "moshi-v1"
+    },
+    {
+      "id": "lambda",
+      "name": "Lambda Labs",
+      "country": "United States",
+      "api_endpoint": "https://api.lambdalabs.com/v1",
+      "api_style": "openai",
+      "supports_model_listing": true,
+      "supports_streaming": true,
+      "supports_vision": false,
+      "supports_embeddings": false,
+      "supports_audio": false,
+      "supports_reasoning": false,
+      "openai_compatible": true,
+      "free_models_available": false,
+      "paid_models_available": true,
+      "needs_key": true,
+      "description": "Lambda Labs — GPU cloud for AI",
+      "last_updated": "2026-09-03T00:00:00+00:00",
+      "fallback_models": [
+        "llama-3.1-405b-instruct-fp8",
+        "llama-3.1-70b-instruct",
+        "llama-3.1-8b-instruct",
+        "hermes-2-pro",
+        "llama3-70b"
+      ],
+      "default_model": "llama-3.1-70b-instruct"
+    },
+    {
+      "id": "lepton",
+      "name": "Lepton AI",
+      "country": "United States",
+      "api_endpoint": "https://lepton.ai/api/v1",
+      "api_style": "openai",
+      "supports_model_listing": true,
+      "supports_streaming": true,
+      "supports_vision": false,
+      "supports_embeddings": false,
+      "supports_audio": false,
+      "supports_reasoning": false,
+      "openai_compatible": true,
+      "free_models_available": true,
+      "paid_models_available": true,
+      "needs_key": true,
+      "description": "Serverless GPU cloud",
+      "last_updated": "2026-07-31T05:17:14+00:00",
+      "fallback_models": [
+        "llama3-70b",
+        "mixtral-8x22b"
+      ],
+      "default_model": "llama3-70b"
+    },
+    {
+      "id": "liquid_ai",
+      "name": "Liquid AI",
+      "country": "United States",
+      "api_endpoint": "https://api.liquid.ai/v1",
+      "api_style": "openai",
+      "supports_model_listing": true,
+      "supports_streaming": true,
+      "supports_vision": false,
+      "supports_embeddings": false,
+      "supports_audio": false,
+      "supports_reasoning": true,
+      "openai_compatible": true,
+      "free_models_available": true,
+      "paid_models_available": true,
+      "needs_key": true,
+      "description": "Liquid Neural Networks (LNN) and Liquid Foundation Models (LFM-40B, LFM-7B)",
+      "last_updated": "2026-09-01T00:00:00+00:00",
+      "fallback_models": [
+        "lfm-40b-chat",
+        "lfm-7b-chat",
+        "lfm-3b-chat"
+      ],
+      "default_model": "lfm-40b-chat"
+    },
+    {
+      "id": "llama-api",
+      "name": "Llama API",
+      "country": "United States",
+      "api_endpoint": "https://api.llama-api.com/v1",
+      "api_style": "openai",
+      "supports_model_listing": true,
+      "supports_streaming": true,
+      "supports_vision": false,
+      "supports_embeddings": false,
+      "supports_audio": false,
+      "supports_reasoning": false,
+      "openai_compatible": true,
+      "free_models_available": true,
+      "paid_models_available": true,
+      "needs_key": true,
+      "description": "Meta Llama hosted API",
+      "last_updated": "2026-07-31T05:17:14+00:00",
+      "fallback_models": [
+        "llama3-70b",
+        "llama3-8b"
+      ],
+      "default_model": "llama3-70b"
+    },
+    {
+      "id": "llamacpp_server",
+      "name": "llama.cpp Server (Local)",
+      "country": "International",
+      "api_endpoint": "http://localhost:8080/v1",
+      "api_style": "openai",
+      "supports_model_listing": true,
+      "supports_streaming": true,
+      "supports_vision": true,
+      "supports_embeddings": true,
+      "supports_audio": false,
+      "supports_reasoning": true,
+      "openai_compatible": true,
+      "free_models_available": true,
+      "paid_models_available": false,
+      "needs_key": false,
+      "description": "Lightweight, dependency-free local inference server running GGUF models on CPU and GPU",
+      "last_updated": "2026-09-01T00:00:00+00:00",
+      "fallback_models": [
+        "gpt-3.5-turbo",
+        "llama-3.1-8b-instruct",
+        "qwen2.5-coder-7b"
+      ],
+      "default_model": "gpt-3.5-turbo"
+    },
+    {
+      "id": "lmstudio",
+      "name": "LM Studio",
+      "country": "International",
+      "api_endpoint": "http://localhost:1234/v1",
+      "api_style": "openai",
+      "supports_model_listing": true,
+      "supports_streaming": true,
+      "supports_vision": false,
+      "supports_embeddings": true,
+      "supports_audio": false,
+      "supports_reasoning": false,
+      "openai_compatible": true,
+      "free_models_available": true,
+      "paid_models_available": false,
+      "needs_key": false,
+      "description": "LM Studio — Local model runner",
+      "last_updated": "2026-09-03T00:00:00+00:00",
+      "fallback_models": [
+        "local-model"
+      ],
+      "default_model": "local-model"
+    },
+    {
+      "id": "lmdeploy_hosted",
+      "name": "LMDeploy Server (Shanghai AI Lab)",
+      "country": "China",
+      "api_endpoint": "http://localhost:23333/v1",
+      "api_style": "openai",
+      "supports_model_listing": true,
+      "supports_streaming": true,
+      "supports_vision": true,
+      "supports_embeddings": false,
+      "supports_audio": false,
+      "supports_reasoning": true,
+      "openai_compatible": true,
+      "free_models_available": true,
+      "paid_models_available": false,
+      "needs_key": false,
+      "description": "High-performance toolkit for compressing, deploying, and serving LLM and VLM models",
+      "last_updated": "2026-09-01T00:00:00+00:00",
+      "fallback_models": [
+        "internlm2_5-20b-chat",
+        "internlm-xcomposer2d5-7b",
+        "qwen2.5-72b-instruct"
+      ],
+      "default_model": "internlm2_5-20b-chat"
+    },
+    {
+      "id": "lmsys",
+      "name": "LMSYS (FastChat)",
+      "country": "United States",
+      "api_endpoint": "https://api.lmsys.org/v1",
+      "api_style": "openai",
+      "supports_model_listing": true,
+      "supports_streaming": true,
+      "supports_vision": false,
+      "supports_embeddings": false,
+      "supports_audio": false,
+      "supports_reasoning": false,
+      "openai_compatible": true,
+      "free_models_available": true,
+      "paid_models_available": false,
+      "needs_key": false,
+      "description": "Open chat model arena",
+      "last_updated": "2026-07-31T05:17:14+00:00",
+      "fallback_models": [
+        "vicuna-13b"
+      ],
+      "default_model": "vicuna-13b"
+    },
+    {
+      "id": "localai",
+      "name": "LocalAI",
+      "country": "International",
+      "api_endpoint": "http://localhost:8080/v1",
+      "api_style": "openai",
+      "supports_model_listing": true,
+      "supports_streaming": true,
+      "supports_vision": false,
+      "supports_embeddings": false,
+      "supports_audio": false,
+      "supports_reasoning": false,
+      "openai_compatible": true,
+      "free_models_available": true,
+      "paid_models_available": false,
+      "needs_key": false,
+      "description": "Self-hosted OpenAI-compatible API",
+      "last_updated": "2026-07-31T05:17:14+00:00",
+      "fallback_models": [
+        "llama3",
+        "phi-3"
+      ],
+      "default_model": "llama3"
+    },
+    {
+      "id": "luma",
+      "name": "Luma Dream Machine",
+      "country": "United States",
+      "api_endpoint": "https://api.lumalabs.ai/v1",
+      "api_style": "openai",
+      "supports_model_listing": true,
+      "supports_streaming": false,
+      "supports_vision": true,
+      "supports_embeddings": false,
+      "supports_audio": false,
+      "supports_reasoning": false,
+      "openai_compatible": false,
+      "free_models_available": false,
+      "paid_models_available": true,
+      "needs_key": true,
+      "description": "Dream Machine next-generation generative video and 3D NeRFs",
+      "last_updated": "2026-09-01T00:00:00+00:00",
+      "fallback_models": [
+        "dream-machine",
+        "ray-1-6"
+      ],
+      "default_model": "dream-machine"
+    },
+    {
+      "id": "maritaca",
+      "name": "Maritaca AI (Sabia)",
+      "country": "Brazil",
+      "api_endpoint": "https://api.maritaca.ai/v1",
+      "api_style": "openai",
+      "supports_model_listing": true,
+      "supports_streaming": true,
+      "supports_vision": false,
+      "supports_embeddings": false,
+      "supports_audio": false,
+      "supports_reasoning": false,
+      "openai_compatible": true,
+      "free_models_available": true,
+      "paid_models_available": true,
+      "needs_key": true,
+      "description": "Portuguese-language LLM",
+      "last_updated": "2026-07-31T05:17:14+00:00",
+      "fallback_models": [
+        "sabia-3"
+      ],
+      "default_model": "sabia-3"
+    },
+    {
+      "id": "microsoft-azure",
+      "name": "Microsoft Azure OpenAI",
+      "country": "United States",
+      "api_endpoint": "https://YOUR_RESOURCE.openai.azure.com/v1",
+      "api_style": "openai",
+      "supports_model_listing": true,
+      "supports_streaming": true,
+      "supports_vision": false,
+      "supports_embeddings": true,
+      "supports_audio": false,
+      "supports_reasoning": false,
+      "openai_compatible": true,
+      "free_models_available": true,
+      "paid_models_available": true,
+      "needs_key": true,
+      "description": "Azure OpenAI Service",
+      "last_updated": "2026-07-31T05:17:14+00:00",
+      "fallback_models": [
+        "gpt-4o",
+        "gpt-4o-mini"
+      ],
+      "default_model": "gpt-4o-mini"
+    },
+    {
+      "id": "milvus",
+      "name": "Milvus (Zilliz Cloud)",
+      "country": "United States",
+      "api_endpoint": "https://controller.api.cloud.zilliz.com/v1",
+      "api_style": "openai",
+      "supports_model_listing": true,
+      "supports_streaming": false,
+      "supports_vision": false,
+      "supports_embeddings": true,
+      "supports_audio": false,
+      "supports_reasoning": false,
+      "openai_compatible": false,
+      "free_models_available": true,
+      "paid_models_available": true,
+      "needs_key": true,
+      "description": "Massive-scale distributed vector database for enterprise AI",
+      "last_updated": "2026-09-01T00:00:00+00:00",
+      "fallback_models": [
+        "bge-m3",
+        "voyage-code-2"
+      ],
+      "default_model": "bge-m3"
+    },
+    {
+      "id": "minimax",
+      "name": "MiniMax",
+      "country": "China",
+      "api_endpoint": "https://api.minimax.chat/v1",
+      "api_style": "openai",
+      "supports_model_listing": true,
+      "supports_streaming": true,
+      "supports_vision": false,
+      "supports_embeddings": false,
+      "supports_audio": false,
+      "supports_reasoning": false,
+      "openai_compatible": true,
+      "free_models_available": true,
+      "paid_models_available": true,
+      "needs_key": true,
+      "description": "MiniMax — Chinese LLM with long context",
+      "last_updated": "2026-09-03T00:00:00+00:00",
+      "fallback_models": [
+        "abab6.5-chat",
+        "abab5.5-chat",
+        "abab4-chat"
+      ],
+      "default_model": "abab6.5-chat"
+    },
+    {
+      "id": "mistral",
+      "name": "Mistral AI",
+      "country": "France",
+      "api_endpoint": "https://api.mistral.ai/v1",
+      "api_style": "openai",
+      "supports_model_listing": true,
+      "supports_streaming": true,
+      "supports_vision": false,
+      "supports_embeddings": true,
+      "supports_audio": false,
+      "supports_reasoning": true,
+      "openai_compatible": true,
+      "free_models_available": true,
+      "paid_models_available": true,
+      "needs_key": true,
+      "description": "Mistral Large & Codestral — European leader",
+      "last_updated": "2026-07-31T05:17:14+00:00",
+      "fallback_models": [
+        "codestral-2405",
+        "mistral-large-2407",
+        "mistral-large-2411",
+        "mistral-large-latest",
+        "mistral-medium",
+        "mistral-medium-latest",
+        "mistral-small-2402",
+        "mistral-small-latest",
+        "mistral-tiny",
+        "open-mistral-nemo",
+        "open-mixtral-8x22b",
+        "open-mixtral-8x7b"
+      ],
+      "default_model": "open-mistral-nemo"
+    },
+    {
+      "id": "modal",
+      "name": "Modal",
+      "country": "United States",
+      "api_endpoint": "https://api.modal.com/v1",
+      "api_style": "openai",
+      "supports_model_listing": true,
+      "supports_streaming": true,
+      "supports_vision": false,
+      "supports_embeddings": false,
+      "supports_audio": false,
+      "supports_reasoning": false,
+      "openai_compatible": true,
+      "free_models_available": true,
+      "paid_models_available": true,
+      "needs_key": true,
+      "description": "Serverless cloud for AI",
+      "last_updated": "2026-07-31T05:17:14+00:00",
+      "fallback_models": [
+        "llama3-70b"
+      ],
+      "default_model": "llama3-70b"
+    },
+    {
+      "id": "monsterapi",
+      "name": "Monster API",
+      "country": "United States",
+      "api_endpoint": "https://api.monsterapi.ai/v1",
+      "api_style": "openai",
+      "supports_model_listing": true,
+      "supports_streaming": true,
+      "supports_vision": true,
+      "supports_embeddings": false,
+      "supports_audio": false,
+      "supports_reasoning": false,
+      "openai_compatible": true,
+      "free_models_available": true,
+      "paid_models_available": true,
+      "needs_key": true,
+      "description": "Affordable model API",
+      "last_updated": "2026-07-31T05:17:14+00:00",
+      "fallback_models": [
+        "llama3-70b",
+        "mixtral-8x7b",
+        "monster-llama-3.1-8b",
+        "monster-mistral-7b",
+        "monster-whisper-large-v3"
+      ],
+      "default_model": "llama3-70b"
+    },
+    {
+      "id": "moonshot",
+      "name": "Moonshot AI (Kimi)",
+      "country": "China",
+      "api_endpoint": "https://api.moonshot.cn/v1",
+      "api_style": "openai",
+      "supports_model_listing": true,
+      "supports_streaming": true,
+      "supports_vision": false,
+      "supports_embeddings": false,
+      "supports_audio": false,
+      "supports_reasoning": true,
+      "openai_compatible": true,
+      "free_models_available": true,
+      "paid_models_available": true,
+      "needs_key": true,
+      "description": "Kimi K3 — latest long-context Chinese LLM",
+      "last_updated": "2026-09-03T00:00:00+00:00",
+      "fallback_models": [
+        "kimi-k3",
+        "moonshot-v1-128k",
+        "moonshot-v1-32k",
+        "moonshot-v1-8k"
+      ],
+      "default_model": "kimi-k3"
+    },
+    {
+      "id": "ncloud",
+      "name": "Naver Cloud (HyperCLOVA X)",
+      "country": "South Korea",
+      "api_endpoint": "https://clovastudio.apigw.ntruss.com/v1",
+      "api_style": "openai",
+      "supports_model_listing": true,
+      "supports_streaming": true,
+      "supports_vision": false,
+      "supports_embeddings": false,
+      "supports_audio": false,
+      "supports_reasoning": false,
+      "openai_compatible": true,
+      "free_models_available": true,
+      "paid_models_available": true,
+      "needs_key": true,
+      "description": "HyperCLOVA X — Naver's enterprise LLM",
+      "last_updated": "2026-09-03T00:00:00+00:00",
+      "fallback_models": [
+        "HCX-005",
+        "HCX-003"
+      ],
+      "default_model": "HCX-005"
+    },
+    {
+      "id": "nebius",
+      "name": "Nebius AI",
+      "country": "Netherlands",
+      "api_endpoint": "https://api.nebius.ai/v1",
+      "api_style": "openai",
+      "supports_model_listing": true,
+      "supports_streaming": true,
+      "supports_vision": false,
+      "supports_embeddings": false,
+      "supports_audio": false,
+      "supports_reasoning": false,
+      "openai_compatible": true,
+      "free_models_available": true,
+      "paid_models_available": true,
+      "needs_key": true,
+      "description": "European GPU cloud",
+      "last_updated": "2026-07-31T05:17:14+00:00",
+      "fallback_models": [
+        "llama3-70b",
+        "qwen2-72b"
+      ],
+      "default_model": "llama3-70b"
+    },
+    {
+      "id": "neets",
+      "name": "Neets AI",
+      "country": "United States",
+      "api_endpoint": "https://api.neets.ai/v1",
+      "api_style": "openai",
+      "supports_model_listing": true,
+      "supports_streaming": true,
+      "supports_vision": false,
+      "supports_embeddings": false,
+      "supports_audio": false,
+      "supports_reasoning": false,
+      "openai_compatible": true,
+      "free_models_available": true,
+      "paid_models_available": true,
+      "needs_key": true,
+      "description": "API gateway with multiple backends",
+      "last_updated": "2026-07-31T05:17:14+00:00",
+      "fallback_models": [
+        "gpt-4o",
+        "llama3-70b"
+      ],
+      "default_model": "gpt-4o"
+    },
+    {
+      "id": "neuralmagic",
+      "name": "Neural Magic",
+      "country": "United States",
+      "api_endpoint": "https://api.neuralmagic.com/v1",
+      "api_style": "openai",
+      "supports_model_listing": true,
+      "supports_streaming": true,
+      "supports_vision": false,
+      "supports_embeddings": false,
+      "supports_audio": false,
+      "supports_reasoning": false,
+      "openai_compatible": true,
+      "free_models_available": true,
+      "paid_models_available": true,
+      "needs_key": true,
+      "description": "CPU-optimized inference",
+      "last_updated": "2026-07-31T05:17:14+00:00",
+      "fallback_models": [
+        "llama3-70b"
+      ],
+      "default_model": "llama3-70b"
+    },
+    {
+      "id": "nomic",
+      "name": "Nomic AI",
+      "country": "United States",
+      "api_endpoint": "https://api.nomic.ai/v1",
+      "api_style": "openai",
+      "supports_model_listing": true,
+      "supports_streaming": false,
+      "supports_vision": false,
+      "supports_embeddings": true,
+      "supports_audio": false,
+      "supports_reasoning": false,
+      "openai_compatible": true,
+      "free_models_available": true,
+      "paid_models_available": true,
+      "needs_key": true,
+      "description": "Embedding models & GPT4All",
+      "last_updated": "2026-07-31T05:17:14+00:00",
+      "fallback_models": [
+        "nomic-embed-text-v1"
+      ],
+      "default_model": "nomic-embed-text-v1"
+    },
+    {
+      "id": "notebooklm",
+      "name": "NotebookLM",
+      "country": "United States",
+      "api_endpoint": "https://notebooklm.google.com",
+      "api_style": "openai",
+      "supports_model_listing": true,
+      "supports_streaming": true,
+      "supports_vision": false,
+      "supports_embeddings": false,
+      "supports_audio": false,
+      "supports_reasoning": false,
+      "openai_compatible": false,
+      "free_models_available": true,
+      "paid_models_available": false,
+      "needs_key": false,
+      "description": "Google research assistant",
+      "last_updated": "2026-07-31T05:17:14+00:00",
+      "fallback_models": [
+        "default"
+      ],
+      "default_model": "default"
+    },
+    {
+      "id": "novita",
+      "name": "Novita AI",
+      "country": "United States",
+      "api_endpoint": "https://api.novita.ai/v3/openai",
+      "api_style": "openai",
+      "supports_model_listing": true,
+      "supports_streaming": true,
+      "supports_vision": false,
+      "supports_embeddings": false,
+      "supports_audio": false,
+      "supports_reasoning": false,
+      "openai_compatible": true,
+      "free_models_available": true,
+      "paid_models_available": true,
+      "needs_key": true,
+      "description": "Novita AI — Cost-effective inference",
+      "last_updated": "2026-09-03T00:00:00+00:00",
+      "fallback_models": [
+        "meta-llama/llama-3.1-70b-instruct",
+        "meta-llama/llama-3.1-8b-instruct",
+        "mistralai/mixtral-8x7b-instruct"
+      ],
+      "default_model": "meta-llama/llama-3.1-70b-instruct"
+    },
+    {
+      "id": "nvidia",
+      "name": "NVIDIA NIM",
+      "country": "United States",
+      "api_endpoint": "https://integrate.api.nvidia.com/v1",
+      "api_style": "openai",
+      "supports_model_listing": true,
+      "supports_streaming": true,
+      "supports_vision": false,
+      "supports_embeddings": false,
+      "supports_audio": false,
+      "supports_reasoning": false,
+      "openai_compatible": true,
+      "free_models_available": true,
+      "paid_models_available": true,
+      "needs_key": true,
+      "description": "NVIDIA Inference Microservices",
+      "last_updated": "2026-07-31T05:17:14+00:00",
+      "fallback_models": [
+        "meta/llama-3.2-11b-vision-instruct",
+        "mistralai/mistral-large-2-instruct",
+        "nvidia/llama-3.1-nemotron-70b-instruct",
+        "ibm/granite-3.0-8b-instruct"
+      ],
+      "default_model": "meta/llama-3.2-11b-vision-instruct"
+    },
+    {
+      "id": "oci-genai",
+      "name": "OCI Generative AI",
+      "country": "United States",
+      "api_endpoint": "https://inference.generativeai.us-chicago-1.oci.oraclecloud.com",
+      "api_style": "openai",
+      "supports_model_listing": true,
+      "supports_streaming": true,
+      "supports_vision": false,
+      "supports_embeddings": false,
+      "supports_audio": false,
+      "supports_reasoning": false,
+      "openai_compatible": true,
+      "free_models_available": true,
+      "paid_models_available": true,
+      "needs_key": true,
+      "description": "Oracle Cloud — managed AI service",
+      "last_updated": "2026-07-31T05:17:14+00:00",
+      "fallback_models": [
+        "cohere.command-r-plus",
+        "meta.llama-3-70b-instruct"
+      ],
+      "default_model": "cohere.command-r-plus"
+    },
+    {
+      "id": "octoai",
+      "name": "OctoAI",
+      "country": "United States",
+      "api_endpoint": "https://text.octoai.run/v1",
+      "api_style": "openai",
+      "supports_model_listing": true,
+      "supports_streaming": true,
+      "supports_vision": true,
+      "supports_embeddings": true,
+      "supports_audio": false,
+      "supports_reasoning": true,
+      "openai_compatible": true,
+      "free_models_available": false,
+      "paid_models_available": true,
+      "needs_key": true,
+      "description": "High-performance inference engine for text, vision, and custom open-source models",
+      "last_updated": "2026-09-01T00:00:00+00:00",
+      "fallback_models": [
+        "meta-llama-3.1-70b-instruct",
+        "mixtral-8x22b-instruct",
+        "hermes-2-pro-mistral-7b"
+      ],
+      "default_model": "meta-llama-3.1-70b-instruct"
+    },
+    {
+      "id": "ollama",
+      "name": "Ollama (Local)",
+      "country": "International",
+      "api_endpoint": "http://localhost:11434",
+      "api_style": "ollama",
+      "supports_model_listing": true,
+      "supports_streaming": true,
+      "supports_vision": false,
+      "supports_embeddings": false,
+      "supports_audio": false,
+      "supports_reasoning": false,
+      "openai_compatible": true,
+      "free_models_available": true,
+      "paid_models_available": false,
+      "needs_key": false,
+      "description": "Local open-source model runner with auto-discovery",
+      "last_updated": "2026-09-03T00:00:00+00:00",
+      "supports_custom_models": true,
+      "auto_discover_models": true,
+      "fallback_models": [
+        "codellama",
+        "deepseek-r1",
+        "gemma2",
+        "gemma2:latest",
+        "llama3",
+        "llama3.1",
+        "llama3.2",
+        "llama3.3",
+        "llama3.3:latest",
+        "mistral",
+        "mixtral",
+        "phi3",
+        "phi4",
+        "qwen2.5",
+        "qwen2.5:latest",
+        "qwen3",
+        "command-r",
+        "dolphin-mistral",
+        "neural-chat",
+        "starling-lm",
+        "vicuna",
+        "yi",
+        "openhermes",
+        "custom-model"
+      ],
+      "default_model": "llama3.3"
+    },
+    {
+      "id": "ollama_cloud",
+      "name": "Ollama Cloud (Hosted)",
+      "country": "International",
+      "api_endpoint": "https://cloud.ollama.ai/v1",
+      "api_style": "ollama",
+      "supports_model_listing": true,
+      "supports_streaming": true,
+      "supports_vision": true,
+      "supports_embeddings": true,
+      "supports_audio": false,
+      "supports_reasoning": true,
+      "openai_compatible": true,
+      "free_models_available": false,
+      "paid_models_available": true,
+      "needs_key": true,
+      "description": "Cloud-managed Ollama instances providing high-availability GPU endpoints",
+      "last_updated": "2026-09-01T00:00:00+00:00",
+      "fallback_models": [
+        "llama3.3:70b",
+        "deepseek-r1:70b",
+        "qwen2.5:72b"
+      ],
+      "default_model": "llama3.3:70b"
+    },
+    {
+      "id": "openai",
+      "name": "OpenAI",
+      "country": "United States",
+      "api_endpoint": "https://api.openai.com/v1",
+      "api_style": "openai",
+      "supports_model_listing": true,
+      "supports_streaming": true,
+      "supports_vision": true,
+      "supports_embeddings": true,
+      "supports_audio": true,
+      "supports_reasoning": true,
+      "openai_compatible": true,
+      "free_models_available": true,
+      "paid_models_available": true,
+      "needs_key": true,
+      "description": "GPT-4o & o1 — industry standard",
+      "last_updated": "2026-09-03T00:00:00+00:00",
+      "fallback_models": [
+        "dall-e-3",
+        "gpt-3.5-turbo",
+        "gpt-4",
+        "gpt-4-turbo",
+        "gpt-4.1",
+        "gpt-4.1-mini",
+        "gpt-4.1-nano",
+        "gpt-4o",
+        "gpt-4o-2024-08-06",
+        "gpt-4o-mini",
+        "gpt-4o-mini-2024-07-18",
+        "o1",
+        "o1-mini",
+        "o3",
+        "o3-mini",
+        "o4-mini",
+        "text-embedding-3-large",
+        "text-embedding-3-small",
+        "tts-1",
+        "whisper-1"
+      ],
+      "default_model": "gpt-4o"
+    },
+    {
+      "id": "openpipe",
+      "name": "OpenPipe",
+      "country": "United States",
+      "api_endpoint": "https://app.openpipe.ai/api/v1",
+      "api_style": "openai",
+      "supports_model_listing": true,
+      "supports_streaming": true,
+      "supports_vision": false,
+      "supports_embeddings": false,
+      "supports_audio": false,
+      "supports_reasoning": true,
+      "openai_compatible": true,
+      "free_models_available": false,
+      "paid_models_available": true,
+      "needs_key": true,
+      "description": "Continuous model fine-tuning, distillation, and optimized inference for production apps",
+      "last_updated": "2026-09-01T00:00:00+00:00",
+      "fallback_models": [
+        "openpipe:llama-3-8b-instruct",
+        "openpipe:mistral-7b-instruct"
+      ],
+      "default_model": "openpipe:llama-3-8b-instruct"
+    },
+    {
+      "id": "openrouter",
+      "name": "OpenRouter",
+      "country": "United States",
+      "api_endpoint": "https://openrouter.ai/api/v1",
+      "api_style": "openai",
+      "supports_model_listing": true,
+      "supports_streaming": true,
+      "supports_vision": true,
+      "supports_embeddings": false,
+      "supports_audio": false,
+      "supports_reasoning": false,
+      "openai_compatible": true,
+      "free_models_available": true,
+      "paid_models_available": true,
+      "needs_key": true,
+      "description": "Multi-provider router up to 200+ models",
+      "last_updated": "2026-07-31T05:17:14+00:00",
+      "fallback_models": [
+        "anthropic/claude-3.5-sonnet",
+        "anthropic/claude-sonnet-4",
+        "deepseek/deepseek-chat",
+        "google/gemini-2.0-flash",
+        "meta-llama/llama-3.1-70b-instruct",
+        "meta-llama/llama-3.3-70b-instruct",
+        "mistralai/mistral-large",
+        "openai/gpt-4o",
+        "openai/gpt-4o-mini"
+      ],
+      "default_model": "google/gemini-2.0-flash"
+    },
+    {
+      "id": "oracle-oci",
+      "name": "Oracle OCI Generative AI",
+      "country": "United States",
+      "api_endpoint": "https://inference.generativeai.us-ashburn-1.oci.oraclecloud.com",
+      "api_style": "openai",
+      "supports_model_listing": true,
+      "supports_streaming": true,
+      "supports_vision": false,
+      "supports_embeddings": false,
+      "supports_audio": false,
+      "supports_reasoning": false,
+      "openai_compatible": true,
+      "free_models_available": false,
+      "paid_models_available": true,
+      "needs_key": true,
+      "description": "Oracle OCI — Enterprise AI inference",
+      "last_updated": "2026-09-03T00:00:00+00:00",
+      "fallback_models": [
+        "meta-llama-3.1-70b-instruct",
+        "meta-llama-3.1-8b-instruct"
+      ],
+      "default_model": "meta-llama-3.1-70b-instruct"
+    },
+    {
+      "id": "ovhcloud",
+      "name": "OVHcloud AI Endpoints",
+      "country": "France",
+      "api_endpoint": "https://endpoints.ai.cloud.ovh.net/v1",
+      "api_style": "openai",
+      "supports_model_listing": true,
+      "supports_streaming": true,
+      "supports_vision": true,
+      "supports_embeddings": true,
+      "supports_audio": false,
+      "supports_reasoning": true,
+      "openai_compatible": true,
+      "free_models_available": true,
+      "paid_models_available": true,
+      "needs_key": true,
+      "description": "European sovereign cloud AI endpoints with complete data privacy and GDPR compliance",
+      "last_updated": "2026-09-01T00:00:00+00:00",
+      "fallback_models": [
+        "Meta-Llama-3-1-70B-Instruct",
+        "Mistral-7B-Instruct-v0.3",
+        "bge-large-en-v1.5"
+      ],
+      "default_model": "Meta-Llama-3-1-70B-Instruct"
+    },
+    {
+      "id": "perplexity",
+      "name": "Perplexity AI",
+      "country": "United States",
+      "api_endpoint": "https://api.perplexity.ai",
+      "api_style": "openai",
+      "supports_model_listing": true,
+      "supports_streaming": true,
+      "supports_vision": false,
+      "supports_embeddings": false,
+      "supports_audio": false,
+      "supports_reasoning": false,
+      "openai_compatible": true,
+      "free_models_available": true,
+      "paid_models_available": true,
+      "needs_key": true,
+      "description": "Sonar — AI search & chat",
+      "last_updated": "2026-07-31T05:17:14+00:00",
+      "fallback_models": [
+        "llama-3.1-sonar-huge-128k",
+        "llama-3.1-sonar-large-128k",
+        "llama-3.1-sonar-small-128k",
+        "sonar",
+        "sonar-pro",
+        "sonar-reasoning",
+        "sonar-reasoning-pro",
+        "sonar-small-chat"
+      ],
+      "default_model": "sonar-reasoning-pro"
+    },
+    {
+      "id": "petals",
+      "name": "Petals",
+      "country": "International",
+      "api_endpoint": "https://petals.dev/v1",
+      "api_style": "openai",
+      "supports_model_listing": true,
+      "supports_streaming": true,
+      "supports_vision": false,
+      "supports_embeddings": false,
+      "supports_audio": false,
+      "supports_reasoning": false,
+      "openai_compatible": true,
+      "free_models_available": true,
+      "paid_models_available": false,
+      "needs_key": false,
+      "description": "Decentralized inference network",
+      "last_updated": "2026-07-31T05:17:14+00:00",
+      "fallback_models": [
+        "llama3-70b"
+      ],
+      "default_model": "llama3-70b"
+    },
+    {
+      "id": "phind",
+      "name": "Phind",
+      "country": "United States",
+      "api_endpoint": "https://api.phind.com/v1",
+      "api_style": "openai",
+      "supports_model_listing": true,
+      "supports_streaming": true,
+      "supports_vision": false,
+      "supports_embeddings": false,
+      "supports_audio": false,
+      "supports_reasoning": false,
+      "openai_compatible": true,
+      "free_models_available": true,
+      "paid_models_available": true,
+      "needs_key": true,
+      "description": "AI search & code assistant",
+      "last_updated": "2026-07-31T05:17:14+00:00",
+      "fallback_models": [
+        "phind-coder",
+        "phind-v2"
+      ],
+      "default_model": "phind-coder"
+    },
+    {
+      "id": "pinecone",
+      "name": "Pinecone AI",
+      "country": "United States",
+      "api_endpoint": "https://api.pinecone.io",
+      "api_style": "openai",
+      "supports_model_listing": true,
+      "supports_streaming": false,
+      "supports_vision": false,
+      "supports_embeddings": true,
+      "supports_audio": false,
+      "supports_reasoning": false,
+      "openai_compatible": false,
+      "free_models_available": true,
+      "paid_models_available": true,
+      "needs_key": true,
+      "description": "Managed vector database and semantic inference service",
+      "last_updated": "2026-09-01T00:00:00+00:00",
+      "fallback_models": [
+        "multilingual-e5-large",
+        "pinecone-rerank-v0"
+      ],
+      "default_model": "multilingual-e5-large"
+    },
+    {
+      "id": "playht",
+      "name": "PlayHT Voice",
+      "country": "United States",
+      "api_endpoint": "https://api.play.ht/api/v2",
+      "api_style": "openai",
+      "supports_model_listing": true,
+      "supports_streaming": true,
+      "supports_vision": false,
+      "supports_embeddings": false,
+      "supports_audio": true,
+      "supports_reasoning": false,
+      "openai_compatible": false,
+      "free_models_available": true,
+      "paid_models_available": true,
+      "needs_key": true,
+      "description": "Conversational AI voice generation and real-time TTS",
+      "last_updated": "2026-09-01T00:00:00+00:00",
+      "fallback_models": [
+        "Play3.0-mini",
+        "PlayDialog"
+      ],
+      "default_model": "Play3.0-mini"
+    },
+    {
+      "id": "poe",
+      "name": "Poe (Quora)",
+      "country": "United States",
+      "api_endpoint": "https://api.poe.com/v1",
+      "api_style": "openai",
+      "supports_model_listing": true,
+      "supports_streaming": true,
+      "supports_vision": true,
+      "supports_embeddings": false,
+      "supports_audio": false,
+      "supports_reasoning": true,
+      "openai_compatible": true,
+      "free_models_available": true,
+      "paid_models_available": true,
+      "needs_key": true,
+      "description": "Quora Poe API protocol for frontier and specialized AI bot interaction",
+      "last_updated": "2026-09-01T00:00:00+00:00",
+      "fallback_models": [
+        "Claude-3.5-Sonnet",
+        "GPT-4o",
+        "DeepSeek-R1"
+      ],
+      "default_model": "Claude-3.5-Sonnet"
+    },
+    {
+      "id": "predibase",
+      "name": "Predibase",
+      "country": "United States",
+      "api_endpoint": "https://api.predibase.com/v1",
+      "api_style": "openai",
+      "supports_model_listing": true,
+      "supports_streaming": true,
+      "supports_vision": false,
+      "supports_embeddings": false,
+      "supports_audio": false,
+      "supports_reasoning": true,
+      "openai_compatible": true,
+      "free_models_available": true,
+      "paid_models_available": true,
+      "needs_key": true,
+      "description": "Serverless fine-tuned & open-weight LLM inference with LoRAX engine",
+      "last_updated": "2026-09-01T00:00:00+00:00",
+      "fallback_models": [
+        "llama-3.3-70b-instruct",
+        "mistral-7b-instruct",
+        "qwen-2.5-coder-32b"
+      ],
+      "default_model": "llama-3.3-70b-instruct"
+    },
+    {
+      "id": "premai",
+      "name": "Prem AI",
+      "country": "United States",
+      "api_endpoint": "https://api.premai.io/v1",
+      "api_style": "openai",
+      "supports_model_listing": true,
+      "supports_streaming": true,
+      "supports_vision": false,
+      "supports_embeddings": false,
+      "supports_audio": false,
+      "supports_reasoning": false,
+      "openai_compatible": true,
+      "free_models_available": true,
+      "paid_models_available": true,
+      "needs_key": true,
+      "description": "Enterprise AI gateway",
+      "last_updated": "2026-07-31T05:17:14+00:00",
+      "fallback_models": [
+        "llama3-70b"
+      ],
+      "default_model": "llama3-70b"
+    },
+    {
+      "id": "qdrant",
+      "name": "Qdrant Vector AI",
+      "country": "Germany",
+      "api_endpoint": "https://api.qdrant.io",
+      "api_style": "openai",
+      "supports_model_listing": true,
+      "supports_streaming": false,
+      "supports_vision": false,
+      "supports_embeddings": true,
+      "supports_audio": false,
+      "supports_reasoning": false,
+      "openai_compatible": false,
+      "free_models_available": true,
+      "paid_models_available": true,
+      "needs_key": true,
+      "description": "High-performance vector search engine and hybrid neural search",
+      "last_updated": "2026-09-01T00:00:00+00:00",
+      "fallback_models": [
+        "bge-large-en-v1.5",
+        "fastembed-hybrid"
+      ],
+      "default_model": "bge-large-en-v1.5"
+    },
+    {
+      "id": "reka",
+      "name": "Reka AI",
+      "country": "United Kingdom",
+      "api_endpoint": "https://api.reka.ai/v1",
+      "api_style": "openai",
+      "supports_model_listing": true,
+      "supports_streaming": true,
+      "supports_vision": true,
+      "supports_embeddings": false,
+      "supports_audio": false,
+      "supports_reasoning": false,
+      "openai_compatible": true,
+      "free_models_available": true,
+      "paid_models_available": true,
+      "needs_key": true,
+      "description": "Reka Core — multimodal AI",
+      "last_updated": "2026-07-31T05:17:14+00:00",
+      "fallback_models": [
+        "reka-core",
+        "reka-flash"
+      ],
+      "default_model": "reka-core"
+    },
+    {
+      "id": "replicate",
+      "name": "Replicate",
+      "country": "United States",
+      "api_endpoint": "https://api.replicate.com/v1",
+      "api_style": "openai",
+      "supports_model_listing": true,
+      "supports_streaming": true,
+      "supports_vision": true,
+      "supports_embeddings": false,
+      "supports_audio": false,
+      "supports_reasoning": false,
+      "openai_compatible": true,
+      "free_models_available": false,
+      "paid_models_available": true,
+      "needs_key": true,
+      "description": "Replicate — Run open-source models",
+      "last_updated": "2026-09-03T00:00:00+00:00",
+      "fallback_models": [
+        "meta/meta-llama-3.1-405b-instruct",
+        "meta/meta-llama-3.1-70b-instruct",
+        "mistralai/mixtral-8x22b-instruct-v0.1"
+      ],
+      "default_model": "meta/meta-llama-3.1-70b-instruct"
+    },
+    {
+      "id": "runpod",
+      "name": "RunPod",
+      "country": "United States",
+      "api_endpoint": "https://api.runpod.ai/v1",
+      "api_style": "openai",
+      "supports_model_listing": true,
+      "supports_streaming": true,
+      "supports_vision": false,
+      "supports_embeddings": false,
+      "supports_audio": false,
+      "supports_reasoning": false,
+      "openai_compatible": true,
+      "free_models_available": true,
+      "paid_models_available": true,
+      "needs_key": true,
+      "description": "Serverless GPU cloud",
+      "last_updated": "2026-07-31T05:17:14+00:00",
+      "fallback_models": [
+        "llama3-70b"
+      ],
+      "default_model": "llama3-70b"
+    },
+    {
+      "id": "runway",
+      "name": "Runway Gen-3",
+      "country": "United States",
+      "api_endpoint": "https://api.runwayml.com/v1",
+      "api_style": "openai",
+      "supports_model_listing": true,
+      "supports_streaming": false,
+      "supports_vision": true,
+      "supports_embeddings": false,
+      "supports_audio": false,
+      "supports_reasoning": false,
+      "openai_compatible": false,
+      "free_models_available": false,
+      "paid_models_available": true,
+      "needs_key": true,
+      "description": "Gen-3 Alpha photorealistic video generation and world models",
+      "last_updated": "2026-09-01T00:00:00+00:00",
+      "fallback_models": [
+        "gen3a_turbo",
+        "gen2"
+      ],
+      "default_model": "gen3a_turbo"
+    },
+    {
+      "id": "sagify",
+      "name": "Sagify AI",
+      "country": "United States",
+      "api_endpoint": "https://api.sagify.ai/v1",
+      "api_style": "openai",
+      "supports_model_listing": true,
+      "supports_streaming": true,
+      "supports_vision": false,
+      "supports_embeddings": false,
+      "supports_audio": false,
+      "supports_reasoning": false,
+      "openai_compatible": true,
+      "free_models_available": true,
+      "paid_models_available": true,
+      "needs_key": true,
+      "description": "ML training & deployment",
+      "last_updated": "2026-07-31T05:17:14+00:00",
+      "fallback_models": [
+        "default"
+      ],
+      "default_model": "default"
+    },
+    {
+      "id": "sakana_ai",
+      "name": "Sakana AI",
+      "country": "Japan",
+      "api_endpoint": "https://api.sakana.ai/v1",
+      "api_style": "openai",
+      "supports_model_listing": true,
+      "supports_streaming": true,
+      "supports_vision": true,
+      "supports_embeddings": false,
+      "supports_audio": false,
+      "supports_reasoning": true,
+      "openai_compatible": true,
+      "free_models_available": false,
+      "paid_models_available": true,
+      "needs_key": true,
+      "description": "Evolutionary model merging and nature-inspired foundation intelligence (EvoLLM, The AI Scientist)",
+      "last_updated": "2026-09-01T00:00:00+00:00",
+      "fallback_models": [
+        "evo-llm-jp",
+        "ai-scientist-v1",
+        "disco-llm"
+      ],
+      "default_model": "evo-llm-jp"
+    },
+    {
+      "id": "sambanova",
+      "name": "SambaNova",
+      "country": "United States",
+      "api_endpoint": "https://api.sambanova.ai/v1",
+      "api_style": "openai",
+      "supports_model_listing": true,
+      "supports_streaming": true,
+      "supports_vision": false,
+      "supports_embeddings": false,
+      "supports_audio": false,
+      "supports_reasoning": false,
+      "openai_compatible": true,
+      "free_models_available": true,
+      "paid_models_available": true,
+      "needs_key": true,
+      "description": "SN40L Reconfigurable Dataflow — fast inference",
+      "last_updated": "2026-07-31T05:17:14+00:00",
+      "fallback_models": [
+        "llama3.1-70b",
+        "llama3.1-8b",
+        "qwen2.5-72b"
+      ],
+      "default_model": "llama3.1-8b"
+    },
+    {
+      "id": "scale_ai",
+      "name": "Scale AI (SEAL)",
+      "country": "United States",
+      "api_endpoint": "https://api.scale.com/v1",
+      "api_style": "openai",
+      "supports_model_listing": true,
+      "supports_streaming": true,
+      "supports_vision": true,
+      "supports_embeddings": true,
+      "supports_audio": false,
+      "supports_reasoning": true,
+      "openai_compatible": true,
+      "free_models_available": false,
+      "paid_models_available": true,
+      "needs_key": true,
+      "description": "Enterprise generative AI platform, frontier model evaluations and SEAL inference",
+      "last_updated": "2026-09-01T00:00:00+00:00",
+      "fallback_models": [
+        "seal-reasoner-v1",
+        "seal-code-70b",
+        "scale-llama-3-70b"
+      ],
+      "default_model": "seal-reasoner-v1"
+    },
+    {
+      "id": "scaleway",
+      "name": "Scaleway Generative APIs",
+      "country": "France",
+      "api_endpoint": "https://api.scaleway.ai/v1",
+      "api_style": "openai",
+      "supports_model_listing": true,
+      "supports_streaming": true,
+      "supports_vision": true,
+      "supports_embeddings": true,
+      "supports_audio": false,
+      "supports_reasoning": true,
+      "openai_compatible": true,
+      "free_models_available": false,
+      "paid_models_available": true,
+      "needs_key": true,
+      "description": "European sovereign cloud infrastructure and hosted open-weight models",
+      "last_updated": "2026-09-01T00:00:00+00:00",
+      "fallback_models": [
+        "llama-3.1-70b-instruct",
+        "mistral-large-2407",
+        "qwen2.5-coder-32b-instruct"
+      ],
+      "default_model": "llama-3.1-70b-instruct"
+    },
+    {
+      "id": "sensetime",
+      "name": "SenseTime (SenseNova)",
+      "country": "China",
+      "api_endpoint": "https://api.sensenova.cn/v1",
+      "api_style": "openai",
+      "supports_model_listing": true,
+      "supports_streaming": true,
+      "supports_vision": true,
+      "supports_embeddings": true,
+      "supports_audio": true,
+      "supports_reasoning": true,
+      "openai_compatible": true,
+      "free_models_available": true,
+      "paid_models_available": true,
+      "needs_key": true,
+      "description": "SenseNova multimodal foundation model family by SenseTime",
+      "last_updated": "2026-09-01T00:00:00+00:00",
+      "fallback_models": [
+        "SenseChat-5",
+        "SenseNova-Omni",
+        "SenseChat-Vision",
+        "nova-5.5-chat",
+        "nova-5-chat",
+        "nova-4-chat"
+      ],
+      "default_model": "SenseChat-5"
+    },
+    {
+      "id": "seven",
+      "name": "Seven AI",
+      "country": "Germany",
+      "api_endpoint": "https://api.seven.ai/v1",
+      "api_style": "openai",
+      "supports_model_listing": true,
+      "supports_streaming": true,
+      "supports_vision": false,
+      "supports_embeddings": false,
+      "supports_audio": false,
+      "supports_reasoning": false,
+      "openai_compatible": true,
+      "free_models_available": true,
+      "paid_models_available": true,
+      "needs_key": true,
+      "description": "German AI platform",
+      "last_updated": "2026-07-31T05:17:14+00:00",
+      "fallback_models": [
+        "default"
+      ],
+      "default_model": "default"
+    },
+    {
+      "id": "sglang_hosted",
+      "name": "SGLang Server (Local / Remote)",
+      "country": "International",
+      "api_endpoint": "http://localhost:30000/v1",
+      "api_style": "openai",
+      "supports_model_listing": true,
+      "supports_streaming": true,
+      "supports_vision": true,
+      "supports_embeddings": false,
+      "supports_audio": false,
+      "supports_reasoning": true,
+      "openai_compatible": true,
+      "free_models_available": true,
+      "paid_models_available": false,
+      "needs_key": false,
+      "description": "Fast serving framework for complex reasoning, multi-turn chat, and structured generation",
+      "last_updated": "2026-09-01T00:00:00+00:00",
+      "fallback_models": [
+        "default",
+        "deepseek-ai/DeepSeek-R1-Distill-Qwen-32B",
+        "meta-llama/Llama-3.3-70B-Instruct"
+      ],
+      "default_model": "default"
+    },
+    {
+      "id": "siliconflow",
+      "name": "SiliconFlow",
+      "country": "China",
+      "api_endpoint": "https://api.siliconflow.cn/v1",
+      "api_style": "openai",
+      "supports_model_listing": true,
+      "supports_streaming": true,
+      "supports_vision": false,
+      "supports_embeddings": false,
+      "supports_audio": false,
+      "supports_reasoning": false,
+      "openai_compatible": true,
+      "free_models_available": true,
+      "paid_models_available": true,
+      "needs_key": true,
+      "description": "Chinese open-model inference platform",
+      "last_updated": "2026-07-31T05:17:14+00:00",
+      "fallback_models": [
+        "deepseek-v3",
+        "glm-4",
+        "qwen2.5-72b"
+      ],
+      "default_model": "deepseek-v3"
+    },
+    {
+      "id": "snowflake",
+      "name": "Snowflake Cortex",
+      "country": "United States",
+      "api_endpoint": "https://<account>.snowflakecomputing.com/api/v2/cortex/chat/completions",
+      "api_style": "openai",
+      "supports_model_listing": true,
+      "supports_streaming": true,
+      "supports_vision": false,
+      "supports_embeddings": false,
+      "supports_audio": false,
+      "supports_reasoning": false,
+      "openai_compatible": true,
+      "free_models_available": false,
+      "paid_models_available": true,
+      "needs_key": true,
+      "description": "Snowflake Cortex — Enterprise data AI",
+      "last_updated": "2026-09-03T00:00:00+00:00",
+      "fallback_models": [
+        "snowflake-arctic",
+        "llama-3.1-70b",
+        "llama-3.1-8b"
+      ],
+      "default_model": "snowflake-arctic"
+    },
+    {
+      "id": "sourcegraph-cody",
+      "name": "Sourcegraph Cody",
+      "country": "United States",
+      "api_endpoint": "https://sourcegraph.com/.api/completions/stream",
+      "api_style": "openai",
+      "supports_model_listing": true,
+      "supports_streaming": true,
+      "supports_vision": false,
+      "supports_embeddings": true,
+      "supports_audio": false,
+      "supports_reasoning": true,
+      "openai_compatible": true,
+      "free_models_available": true,
+      "paid_models_available": true,
+      "needs_key": true,
+      "description": "Code intelligence and multi-repo context assistant",
+      "last_updated": "2026-09-01T00:00:00+00:00",
+      "fallback_models": [
+        "cody-chat",
+        "claude-3.5-sonnet",
+        "gemini-1.5-pro"
+      ],
+      "default_model": "cody-chat"
+    },
+    {
+      "id": "spark",
+      "name": "Spark AI (iFlytek)",
+      "country": "China",
+      "api_endpoint": "https://spark-api.xf-yun.com/v1",
+      "api_style": "openai",
+      "supports_model_listing": true,
+      "supports_streaming": true,
+      "supports_vision": false,
+      "supports_embeddings": false,
+      "supports_audio": true,
+      "supports_reasoning": false,
+      "openai_compatible": true,
+      "free_models_available": true,
+      "paid_models_available": true,
+      "needs_key": true,
+      "description": "iFlytek Spark — Chinese LLM",
+      "last_updated": "2026-07-31T05:17:14+00:00",
+      "fallback_models": [
+        "spark-3.5",
+        "spark-4.0"
+      ],
+      "default_model": "spark-3.5"
+    },
+    {
+      "id": "stability",
+      "name": "Stability AI",
+      "country": "United Kingdom",
+      "api_endpoint": "https://api.stability.ai/v1",
+      "api_style": "openai",
+      "supports_model_listing": true,
+      "supports_streaming": false,
+      "supports_vision": true,
+      "supports_embeddings": false,
+      "supports_audio": false,
+      "supports_reasoning": false,
+      "openai_compatible": false,
+      "free_models_available": true,
+      "paid_models_available": true,
+      "needs_key": true,
+      "description": "Stable Diffusion — image generation",
+      "last_updated": "2026-07-31T05:17:14+00:00",
+      "fallback_models": [
+        "stable-diffusion-3.5",
+        "stable-image-ultra"
+      ],
+      "default_model": "stable-diffusion-3.5"
+    },
+    {
+      "id": "crfm_helm",
+      "name": "Stanford CRFM (HELM)",
+      "country": "United States",
+      "api_endpoint": "https://crfm.stanford.edu/helm/api/v1",
+      "api_style": "openai",
+      "supports_model_listing": true,
+      "supports_streaming": false,
+      "supports_vision": true,
+      "supports_embeddings": true,
+      "supports_audio": false,
+      "supports_reasoning": true,
+      "openai_compatible": true,
+      "free_models_available": true,
+      "paid_models_available": false,
+      "needs_key": true,
+      "description": "Academic evaluation benchmark and model exploration API from Stanford University",
+      "last_updated": "2026-09-01T00:00:00+00:00",
+      "fallback_models": [
+        "helm-eval-suite",
+        "stanford-crfm/llama-2-7b"
+      ],
+      "default_model": "helm-eval-suite"
+    },
+    {
+      "id": "stepfun",
+      "name": "StepFun (Step)",
+      "country": "China",
+      "api_endpoint": "https://api.stepfun.com/v1",
+      "api_style": "openai",
+      "supports_model_listing": true,
+      "supports_streaming": true,
+      "supports_vision": true,
+      "supports_embeddings": false,
+      "supports_audio": false,
+      "supports_reasoning": false,
+      "openai_compatible": true,
+      "free_models_available": true,
+      "paid_models_available": true,
+      "needs_key": true,
+      "description": "Step 2 — StepFun's long-context LLM",
+      "last_updated": "2026-09-03T00:00:00+00:00",
+      "fallback_models": [
+        "step-2-16k",
+        "step-2-32k",
+        "step-1-flash"
+      ],
+      "default_model": "step-2-16k"
+    },
+    {
+      "id": "stochastic",
+      "name": "Stochastic AI",
+      "country": "United States",
+      "api_endpoint": "https://api.stochastic.ai/v1",
+      "api_style": "openai",
+      "supports_model_listing": true,
+      "supports_streaming": true,
+      "supports_vision": false,
+      "supports_embeddings": false,
+      "supports_audio": false,
+      "supports_reasoning": false,
+      "openai_compatible": true,
+      "free_models_available": true,
+      "paid_models_available": true,
+      "needs_key": true,
+      "description": "Model inference API",
+      "last_updated": "2026-07-31T05:17:14+00:00",
+      "fallback_models": [
+        "llama3-70b"
+      ],
+      "default_model": "llama3-70b"
+    },
+    {
+      "id": "suno",
+      "name": "Suno AI Music",
+      "country": "United States",
+      "api_endpoint": "https://api.suno.ai/v1",
+      "api_style": "openai",
+      "supports_model_listing": true,
+      "supports_streaming": true,
+      "supports_vision": false,
+      "supports_embeddings": false,
+      "supports_audio": true,
+      "supports_reasoning": false,
+      "openai_compatible": false,
+      "free_models_available": false,
+      "paid_models_available": true,
+      "needs_key": true,
+      "description": "Generative musical composition, arrangement, and vocal synthesis",
+      "last_updated": "2026-09-01T00:00:00+00:00",
+      "fallback_models": [
+        "chirp-v3-5",
+        "chirp-v3"
+      ],
+      "default_model": "chirp-v3-5"
+    },
+    {
+      "id": "tabby_api",
+      "name": "TabbyAPI (ExLlamaV2)",
+      "country": "International",
+      "api_endpoint": "http://localhost:5000/v1",
+      "api_style": "openai",
+      "supports_model_listing": true,
+      "supports_streaming": true,
+      "supports_vision": false,
+      "supports_embeddings": false,
+      "supports_audio": false,
+      "supports_reasoning": true,
+      "openai_compatible": true,
+      "free_models_available": true,
+      "paid_models_available": false,
+      "needs_key": true,
+      "description": "Fast local inference server hosting EXL2 quantized models with OpenAI-compatible API",
+      "last_updated": "2026-09-01T00:00:00+00:00",
+      "fallback_models": [
+        "current",
+        "llama-3.3-70b-instruct-exl2",
+        "mistral-large-2407-exl2"
+      ],
+      "default_model": "current"
+    },
+    {
+      "id": "tabnine",
+      "name": "Tabnine AI",
+      "country": "Israel",
+      "api_endpoint": "https://api.tabnine.com/v1",
+      "api_style": "openai",
+      "supports_model_listing": true,
+      "supports_streaming": true,
+      "supports_vision": false,
+      "supports_embeddings": false,
+      "supports_audio": false,
+      "supports_reasoning": false,
+      "openai_compatible": true,
+      "free_models_available": true,
+      "paid_models_available": true,
+      "needs_key": true,
+      "description": "Private, secure code completion and generative chat",
+      "last_updated": "2026-09-01T00:00:00+00:00",
+      "fallback_models": [
+        "tabnine-chat",
+        "tabnine-protected"
+      ],
+      "default_model": "tabnine-chat"
+    },
+    {
+      "id": "tencent",
+      "name": "Tencent (Hunyuan)",
+      "country": "China",
+      "api_endpoint": "https://api.hunyuan.cloud.tencent.com/v1",
+      "api_style": "openai",
+      "supports_model_listing": true,
+      "supports_streaming": true,
+      "supports_vision": false,
+      "supports_embeddings": false,
+      "supports_audio": false,
+      "supports_reasoning": false,
+      "openai_compatible": true,
+      "free_models_available": true,
+      "paid_models_available": true,
+      "needs_key": true,
+      "description": "Hunyuan — Tencent's LLM",
+      "last_updated": "2026-07-31T05:17:14+00:00",
+      "fallback_models": [
+        "hunyuan-pro",
+        "hunyuan-standard"
+      ],
+      "default_model": "hunyuan-pro"
+    },
+    {
+      "id": "tencent-hunyuan",
+      "name": "Tencent Hunyuan",
+      "country": "China",
+      "api_endpoint": "https://hunyuan.tencentcloudapi.com",
+      "api_style": "openai",
+      "supports_model_listing": true,
+      "supports_streaming": true,
+      "supports_vision": true,
+      "supports_embeddings": false,
+      "supports_audio": false,
+      "supports_reasoning": false,
+      "openai_compatible": true,
+      "free_models_available": true,
+      "paid_models_available": true,
+      "needs_key": true,
+      "description": "Hunyuan — Tencent's flagship LLM",
+      "last_updated": "2026-09-03T00:00:00+00:00",
+      "fallback_models": [
+        "hunyuan-pro",
+        "hunyuan-standard",
+        "hunyuan-lite"
+      ],
+      "default_model": "hunyuan-pro"
+    },
+    {
+      "id": "text-generation-webui",
+      "name": "text-generation-webui",
+      "country": "International",
+      "api_endpoint": "http://localhost:5000/v1",
+      "api_style": "openai",
+      "supports_model_listing": true,
+      "supports_streaming": true,
+      "supports_vision": false,
+      "supports_embeddings": false,
+      "supports_audio": false,
+      "supports_reasoning": false,
+      "openai_compatible": true,
+      "free_models_available": true,
+      "paid_models_available": false,
+      "needs_key": false,
+      "description": "text-generation-webui — Gradio UI for LLMs",
+      "last_updated": "2026-09-03T00:00:00+00:00",
+      "fallback_models": [
+        "default"
+      ],
+      "default_model": "default"
+    },
+    {
+      "id": "titan",
+      "name": "Titan AI",
+      "country": "United States",
+      "api_endpoint": "https://api.titanai.io/v1",
+      "api_style": "openai",
+      "supports_model_listing": true,
+      "supports_streaming": true,
+      "supports_vision": false,
+      "supports_embeddings": false,
+      "supports_audio": false,
+      "supports_reasoning": false,
+      "openai_compatible": true,
+      "free_models_available": true,
+      "paid_models_available": true,
+      "needs_key": true,
+      "description": "AI platform",
+      "last_updated": "2026-07-31T05:17:14+00:00",
+      "fallback_models": [
+        "default"
+      ],
+      "default_model": "default"
+    },
+    {
+      "id": "together",
+      "name": "Together AI",
+      "country": "United States",
+      "api_endpoint": "https://api.together.xyz/v1",
+      "api_style": "openai",
+      "supports_model_listing": true,
+      "supports_streaming": true,
+      "supports_vision": true,
+      "supports_embeddings": false,
+      "supports_audio": false,
+      "supports_reasoning": false,
+      "openai_compatible": true,
+      "free_models_available": true,
+      "paid_models_available": true,
+      "needs_key": true,
+      "description": "Hosted open-source model inference",
+      "last_updated": "2026-07-31T05:17:14+00:00",
+      "fallback_models": [
+        "deepseek-ai/DeepSeek-V3",
+        "google/gemma-2-27b-it",
+        "meta-llama/Llama-3.1-8B-Instruct-Turbo",
+        "meta-llama/Llama-3.3-70B-Instruct-Turbo",
+        "mistralai/Mixtral-8x22B-Instruct-v0.1",
+        "meta-llama/Llama-3.1-70B-Instruct-Turbo",
+        "deepseek-ai/DeepSeek-R1"
+      ],
+      "default_model": "meta-llama/Llama-3.3-70B-Instruct-Turbo"
+    },
+    {
+      "id": "trulens",
+      "name": "TruLens AI",
+      "country": "United States",
+      "api_endpoint": "https://api.trulens.ai/v1",
+      "api_style": "openai",
+      "supports_model_listing": true,
+      "supports_streaming": true,
+      "supports_vision": false,
+      "supports_embeddings": false,
+      "supports_audio": false,
+      "supports_reasoning": false,
+      "openai_compatible": true,
+      "free_models_available": true,
+      "paid_models_available": true,
+      "needs_key": true,
+      "description": "AI evaluation & monitoring",
+      "last_updated": "2026-07-31T05:17:14+00:00",
+      "fallback_models": [
+        "default"
+      ],
+      "default_model": "default"
+    },
+    {
+      "id": "upstage",
+      "name": "Upstage (Solar)",
+      "country": "South Korea",
+      "api_endpoint": "https://api.upstage.ai/v1/solar",
+      "api_style": "openai",
+      "supports_model_listing": true,
+      "supports_streaming": true,
+      "supports_vision": false,
+      "supports_embeddings": true,
+      "supports_audio": false,
+      "supports_reasoning": false,
+      "openai_compatible": true,
+      "free_models_available": true,
+      "paid_models_available": true,
+      "needs_key": true,
+      "description": "Solar 10.7B — Upstage's efficient LLM",
+      "last_updated": "2026-09-03T00:00:00+00:00",
+      "fallback_models": [
+        "solar-10.7b-instruct",
+        "solar-1-mini",
+        "embedding-passage"
+      ],
+      "default_model": "solar-10.7b-instruct"
+    },
+    {
+      "id": "vllm",
+      "name": "vLLM",
+      "country": "International",
+      "api_endpoint": "http://localhost:8000/v1",
+      "api_style": "openai",
+      "supports_model_listing": true,
+      "supports_streaming": true,
+      "supports_vision": false,
+      "supports_embeddings": false,
+      "supports_audio": false,
+      "supports_reasoning": false,
+      "openai_compatible": true,
+      "free_models_available": true,
+      "paid_models_available": false,
+      "needs_key": false,
+      "description": "Self-hosted high-throughput inference",
+      "last_updated": "2026-07-31T05:17:14+00:00",
+      "fallback_models": [
+        "llama3",
+        "qwen2",
+        "default",
+        "meta-llama/Llama-3.1-8B-Instruct",
+        "deepseek-ai/DeepSeek-R1-Distill-Qwen-14B"
+      ],
+      "default_model": "llama3"
+    },
+    {
+      "id": "voiceflow",
+      "name": "Voiceflow",
+      "country": "United States",
+      "api_endpoint": "https://api.voiceflow.com/v1",
+      "api_style": "openai",
+      "supports_model_listing": true,
+      "supports_streaming": true,
+      "supports_vision": false,
+      "supports_embeddings": false,
+      "supports_audio": false,
+      "supports_reasoning": false,
+      "openai_compatible": true,
+      "free_models_available": true,
+      "paid_models_available": true,
+      "needs_key": true,
+      "description": "Conversational AI platform",
+      "last_updated": "2026-07-31T05:17:14+00:00",
+      "fallback_models": [
+        "default"
+      ],
+      "default_model": "default"
+    },
+    {
+      "id": "voyage",
+      "name": "Voyage AI",
+      "country": "United States",
+      "api_endpoint": "https://api.voyageai.com/v1",
+      "api_style": "openai",
+      "supports_model_listing": true,
+      "supports_streaming": false,
+      "supports_vision": false,
+      "supports_embeddings": true,
+      "supports_audio": false,
+      "supports_reasoning": false,
+      "openai_compatible": true,
+      "free_models_available": true,
+      "paid_models_available": true,
+      "needs_key": true,
+      "description": "State-of-the-art embedding models",
+      "last_updated": "2026-07-31T05:17:14+00:00",
+      "fallback_models": [
+        "voyage-3",
+        "voyage-code-3"
+      ],
+      "default_model": "voyage-3"
+    },
+    {
+      "id": "weaviate",
+      "name": "Weaviate Cloud AI",
+      "country": "Netherlands",
+      "api_endpoint": "https://api.weaviate.io/v1",
+      "api_style": "openai",
+      "supports_model_listing": true,
+      "supports_streaming": false,
+      "supports_vision": true,
+      "supports_embeddings": true,
+      "supports_audio": false,
+      "supports_reasoning": false,
+      "openai_compatible": false,
+      "free_models_available": true,
+      "paid_models_available": true,
+      "needs_key": true,
+      "description": "AI-native vector search engine with generative search modules",
+      "last_updated": "2026-09-01T00:00:00+00:00",
+      "fallback_models": [
+        "text2vec-openai",
+        "generative-cohere"
+      ],
+      "default_model": "text2vec-openai"
+    },
+    {
+      "id": "writer",
+      "name": "Writer AI",
+      "country": "United States",
+      "api_endpoint": "https://api.writer.com/v1",
+      "api_style": "openai",
+      "supports_model_listing": true,
+      "supports_streaming": true,
+      "supports_vision": false,
+      "supports_embeddings": false,
+      "supports_audio": false,
+      "supports_reasoning": false,
+      "openai_compatible": true,
+      "free_models_available": true,
+      "paid_models_available": true,
+      "needs_key": true,
+      "description": "Palmyra — enterprise LLM",
+      "last_updated": "2026-07-31T05:17:14+00:00",
+      "fallback_models": [
+        "palmyra-large",
+        "palmyra-x-004"
+      ],
+      "default_model": "palmyra-large"
+    },
+    {
+      "id": "xai",
+      "name": "xAI (Grok)",
+      "country": "United States",
+      "api_endpoint": "https://api.x.ai/v1",
+      "api_style": "openai",
+      "supports_model_listing": true,
+      "supports_streaming": true,
+      "supports_vision": true,
+      "supports_embeddings": false,
+      "supports_audio": false,
+      "supports_reasoning": true,
+      "openai_compatible": true,
+      "free_models_available": true,
+      "paid_models_available": true,
+      "needs_key": true,
+      "description": "Grok-3 & GLM — xAI's flagship models",
+      "last_updated": "2026-09-03T00:00:00+00:00",
+      "fallback_models": [
+        "grok-2-1212",
+        "grok-2-latest",
+        "grok-2-vision-1212",
+        "grok-3",
+        "grok-3-mini",
+        "grok-beta",
+        "glm-5.2",
+        "glm-5.3"
+      ],
+      "default_model": "grok-3"
+    },
+    {
+      "id": "xenova",
+      "name": "Xenova (Transformers.js)",
+      "country": "Canada",
+      "api_endpoint": "https://api.xenova.ai/v1",
+      "api_style": "openai",
+      "supports_model_listing": true,
+      "supports_streaming": true,
+      "supports_vision": false,
+      "supports_embeddings": false,
+      "supports_audio": false,
+      "supports_reasoning": false,
+      "openai_compatible": true,
+      "free_models_available": true,
+      "paid_models_available": false,
+      "needs_key": false,
+      "description": "Browser/Edge inference via Transformers.js",
+      "last_updated": "2026-07-31T05:17:14+00:00",
+      "fallback_models": [
+        "llama-3.2-1b",
+        "phi-3-mini"
+      ],
+      "default_model": "llama-3.2-1b"
+    },
+    {
+      "id": "zhipu",
+      "name": "Zhipu AI (GLM)",
+      "country": "China",
+      "api_endpoint": "https://open.bigmodel.cn/api/paas/v4",
+      "api_style": "openai",
+      "supports_model_listing": true,
+      "supports_streaming": true,
+      "supports_vision": true,
+      "supports_embeddings": false,
+      "supports_audio": false,
+      "supports_reasoning": false,
+      "openai_compatible": true,
+      "free_models_available": true,
+      "paid_models_available": true,
+      "needs_key": true,
+      "description": "GLM-4 — Zhipu's multilingual LLM",
+      "last_updated": "2026-07-31T05:17:14+00:00",
+      "fallback_models": [
+        "glm-4-air",
+        "glm-4-plus",
+        "glm-4v"
+      ],
+      "default_model": "glm-4-air"
+    },
+    {
+      "id": "zilliz",
+      "name": "Zilliz Cloud",
+      "country": "United States",
+      "api_endpoint": "https://api.zilliz.com/v1",
+      "api_style": "openai",
+      "supports_model_listing": true,
+      "supports_streaming": false,
+      "supports_vision": false,
+      "supports_embeddings": true,
+      "supports_audio": false,
+      "supports_reasoning": false,
+      "openai_compatible": true,
+      "free_models_available": true,
+      "paid_models_available": true,
+      "needs_key": true,
+      "description": "Vector database & embedding API",
+      "last_updated": "2026-09-03T00:00:00+00:00",
+      "fallback_models": [
+        "default"
+      ],
+      "default_model": "default"
+    }
+  ]
+}