sandboxy 0.0.1__py3-none-any.whl

sandboxy/providers/openrouter.py

@@ -0,0 +1,958 @@
+"""OpenRouter provider - unified API for 400+ models."""
+
+import json
+import logging
+import os
+import time
+from collections.abc import AsyncIterator
+from typing import Any
+
+import httpx
+
+logger = logging.getLogger(__name__)
+
+from sandboxy.providers.base import BaseProvider, ModelInfo, ModelResponse, ProviderError
+
+# Popular models with their metadata (subset - OpenRouter has 400+)
+OPENROUTER_MODELS = {
+    # ==========================================================================
+    # OpenAI (Latest: GPT-5.x series)
+    # ==========================================================================
+    "openai/gpt-5.2-pro": ModelInfo(
+        id="openai/gpt-5.2-pro",
+        name="GPT-5.2 Pro",
+        provider="openai",
+        context_length=400000,
+        input_cost_per_million=21.00,
+        output_cost_per_million=168.00,
+        supports_vision=True,
+    ),
+    "openai/gpt-5.2": ModelInfo(
+        id="openai/gpt-5.2",
+        name="GPT-5.2",
+        provider="openai",
+        context_length=400000,
+        input_cost_per_million=1.75,
+        output_cost_per_million=14.00,
+        supports_vision=True,
+    ),
+    "openai/gpt-5.2-codex": ModelInfo(
+        id="openai/gpt-5.2-codex",
+        name="GPT-5.2 Codex",
+        provider="openai",
+        context_length=400000,
+        input_cost_per_million=1.75,
+        output_cost_per_million=14.00,
+        supports_vision=True,
+    ),
+    "openai/gpt-5-mini": ModelInfo(
+        id="openai/gpt-5-mini",
+        name="GPT-5 Mini",
+        provider="openai",
+        context_length=128000,
+        input_cost_per_million=0.30,
+        output_cost_per_million=2.40,
+        supports_vision=True,
+    ),
+    "openai/gpt-5.1": ModelInfo(
+        id="openai/gpt-5.1",
+        name="GPT-5.1",
+        provider="openai",
+        context_length=400000,
+        input_cost_per_million=1.25,
+        output_cost_per_million=10.00,
+        supports_vision=True,
+    ),
+    "openai/gpt-5.1-codex": ModelInfo(
+        id="openai/gpt-5.1-codex",
+        name="GPT-5.1 Codex",
+        provider="openai",
+        context_length=400000,
+        input_cost_per_million=0.30,
+        output_cost_per_million=2.40,
+        supports_vision=True,
+    ),
+    "openai/gpt-4.1": ModelInfo(
+        id="openai/gpt-4.1",
+        name="GPT-4.1",
+        provider="openai",
+        context_length=1000000,
+        input_cost_per_million=2.00,
+        output_cost_per_million=8.00,
+        supports_vision=True,
+    ),
+    "openai/gpt-4.1-mini": ModelInfo(
+        id="openai/gpt-4.1-mini",
+        name="GPT-4.1 Mini",
+        provider="openai",
+        context_length=1000000,
+        input_cost_per_million=0.40,
+        output_cost_per_million=1.60,
+        supports_vision=True,
+    ),
+    "openai/gpt-4.1-nano": ModelInfo(
+        id="openai/gpt-4.1-nano",
+        name="GPT-4.1 Nano",
+        provider="openai",
+        context_length=1000000,
+        input_cost_per_million=0.10,
+        output_cost_per_million=0.40,
+        supports_vision=True,
+    ),
+    "openai/gpt-4o": ModelInfo(
+        id="openai/gpt-4o",
+        name="GPT-4o",
+        provider="openai",
+        context_length=128000,
+        input_cost_per_million=2.50,
+        output_cost_per_million=10.00,
+        supports_vision=True,
+    ),
+    "openai/gpt-4o-mini": ModelInfo(
+        id="openai/gpt-4o-mini",
+        name="GPT-4o Mini",
+        provider="openai",
+        context_length=128000,
+        input_cost_per_million=0.15,
+        output_cost_per_million=0.60,
+        supports_vision=True,
+    ),
+    "openai/o3": ModelInfo(
+        id="openai/o3",
+        name="o3",
+        provider="openai",
+        context_length=200000,
+        input_cost_per_million=20.00,
+        output_cost_per_million=80.00,
+    ),
+    "openai/o3-mini": ModelInfo(
+        id="openai/o3-mini",
+        name="o3 Mini",
+        provider="openai",
+        context_length=200000,
+        input_cost_per_million=1.10,
+        output_cost_per_million=4.40,
+    ),
+    "openai/o1": ModelInfo(
+        id="openai/o1",
+        name="o1",
+        provider="openai",
+        context_length=200000,
+        input_cost_per_million=15.00,
+        output_cost_per_million=60.00,
+    ),
+    "openai/o1-mini": ModelInfo(
+        id="openai/o1-mini",
+        name="o1 Mini",
+        provider="openai",
+        context_length=128000,
+        input_cost_per_million=3.00,
+        output_cost_per_million=12.00,
+    ),
+    "openai/o1-pro": ModelInfo(
+        id="openai/o1-pro",
+        name="o1 Pro",
+        provider="openai",
+        context_length=200000,
+        input_cost_per_million=150.00,
+        output_cost_per_million=600.00,
+    ),
+    # ==========================================================================
+    # Anthropic (Latest: Claude 4.5 series)
+    # ==========================================================================
+    "anthropic/claude-opus-4.5": ModelInfo(
+        id="anthropic/claude-opus-4.5",
+        name="Claude Opus 4.5",
+        provider="anthropic",
+        context_length=200000,
+        input_cost_per_million=5.00,
+        output_cost_per_million=25.00,
+        supports_vision=True,
+    ),
+    "anthropic/claude-sonnet-4.5": ModelInfo(
+        id="anthropic/claude-sonnet-4.5",
+        name="Claude Sonnet 4.5",
+        provider="anthropic",
+        context_length=1000000,
+        input_cost_per_million=3.00,
+        output_cost_per_million=15.00,
+        supports_vision=True,
+    ),
+    "anthropic/claude-haiku-4.5": ModelInfo(
+        id="anthropic/claude-haiku-4.5",
+        name="Claude Haiku 4.5",
+        provider="anthropic",
+        context_length=200000,
+        input_cost_per_million=1.00,
+        output_cost_per_million=5.00,
+        supports_vision=True,
+    ),
+    "anthropic/claude-sonnet-4": ModelInfo(
+        id="anthropic/claude-sonnet-4",
+        name="Claude Sonnet 4",
+        provider="anthropic",
+        context_length=200000,
+        input_cost_per_million=3.00,
+        output_cost_per_million=15.00,
+        supports_vision=True,
+    ),
+    "anthropic/claude-opus-4": ModelInfo(
+        id="anthropic/claude-opus-4",
+        name="Claude Opus 4",
+        provider="anthropic",
+        context_length=200000,
+        input_cost_per_million=15.00,
+        output_cost_per_million=75.00,
+        supports_vision=True,
+    ),
+    "anthropic/claude-3.5-sonnet": ModelInfo(
+        id="anthropic/claude-3.5-sonnet",
+        name="Claude 3.5 Sonnet",
+        provider="anthropic",
+        context_length=200000,
+        input_cost_per_million=3.00,
+        output_cost_per_million=15.00,
+        supports_vision=True,
+    ),
+    "anthropic/claude-3.5-haiku": ModelInfo(
+        id="anthropic/claude-3.5-haiku",
+        name="Claude 3.5 Haiku",
+        provider="anthropic",
+        context_length=200000,
+        input_cost_per_million=0.80,
+        output_cost_per_million=4.00,
+        supports_vision=True,
+    ),
+    "anthropic/claude-3-opus": ModelInfo(
+        id="anthropic/claude-3-opus",
+        name="Claude 3 Opus",
+        provider="anthropic",
+        context_length=200000,
+        input_cost_per_million=15.00,
+        output_cost_per_million=75.00,
+        supports_vision=True,
+    ),
+    "anthropic/claude-3-haiku": ModelInfo(
+        id="anthropic/claude-3-haiku",
+        name="Claude 3 Haiku",
+        provider="anthropic",
+        context_length=200000,
+        input_cost_per_million=0.25,
+        output_cost_per_million=1.25,
+        supports_vision=True,
+    ),
+    # ==========================================================================
+    # Google (Latest: Gemini 3.x series)
+    # ==========================================================================
+    "google/gemini-3-pro": ModelInfo(
+        id="google/gemini-3-pro",
+        name="Gemini 3 Pro",
+        provider="google",
+        context_length=1048576,
+        input_cost_per_million=2.00,
+        output_cost_per_million=12.00,
+        supports_vision=True,
+    ),
+    "google/gemini-3-flash": ModelInfo(
+        id="google/gemini-3-flash",
+        name="Gemini 3 Flash",
+        provider="google",
+        context_length=1048576,
+        input_cost_per_million=0.30,
+        output_cost_per_million=1.20,
+        supports_vision=True,
+    ),
+    "google/gemini-2.5-pro": ModelInfo(
+        id="google/gemini-2.5-pro",
+        name="Gemini 2.5 Pro",
+        provider="google",
+        context_length=1048576,
+        input_cost_per_million=1.25,
+        output_cost_per_million=10.00,
+        supports_vision=True,
+    ),
+    "google/gemini-2.5-flash": ModelInfo(
+        id="google/gemini-2.5-flash",
+        name="Gemini 2.5 Flash",
+        provider="google",
+        context_length=1048576,
+        input_cost_per_million=0.30,
+        output_cost_per_million=2.50,
+        supports_vision=True,
+    ),
+    "google/gemini-2.0-flash": ModelInfo(
+        id="google/gemini-2.0-flash",
+        name="Gemini 2.0 Flash",
+        provider="google",
+        context_length=1000000,
+        input_cost_per_million=0.10,
+        output_cost_per_million=0.40,
+        supports_vision=True,
+    ),
+    "google/gemini-2.0-flash-exp:free": ModelInfo(
+        id="google/gemini-2.0-flash-exp:free",
+        name="Gemini 2.0 Flash (Free)",
+        provider="google",
+        context_length=1000000,
+        input_cost_per_million=0.0,
+        output_cost_per_million=0.0,
+        supports_vision=True,
+    ),
+    "google/gemini-2.0-flash-thinking-exp:free": ModelInfo(
+        id="google/gemini-2.0-flash-thinking-exp:free",
+        name="Gemini 2.0 Flash Thinking (Free)",
+        provider="google",
+        context_length=1000000,
+        input_cost_per_million=0.0,
+        output_cost_per_million=0.0,
+        supports_vision=True,
+    ),
+    "google/gemini-pro-1.5": ModelInfo(
+        id="google/gemini-pro-1.5",
+        name="Gemini Pro 1.5",
+        provider="google",
+        context_length=2000000,
+        input_cost_per_million=1.25,
+        output_cost_per_million=5.00,
+        supports_vision=True,
+    ),
+    "google/gemini-flash-1.5": ModelInfo(
+        id="google/gemini-flash-1.5",
+        name="Gemini Flash 1.5",
+        provider="google",
+        context_length=1000000,
+        input_cost_per_million=0.075,
+        output_cost_per_million=0.30,
+        supports_vision=True,
+    ),
+    # ==========================================================================
+    # xAI (Latest: Grok 4)
+    # ==========================================================================
+    "x-ai/grok-4": ModelInfo(
+        id="x-ai/grok-4",
+        name="Grok 4",
+        provider="xai",
+        context_length=2000000,
+        input_cost_per_million=3.00,
+        output_cost_per_million=15.00,
+        supports_vision=True,
+    ),
+    "x-ai/grok-4-fast": ModelInfo(
+        id="x-ai/grok-4-fast",
+        name="Grok 4 Fast",
+        provider="xai",
+        context_length=2000000,
+        input_cost_per_million=0.20,
+        output_cost_per_million=0.50,
+        supports_vision=True,
+    ),
+    "x-ai/grok-3": ModelInfo(
+        id="x-ai/grok-3",
+        name="Grok 3",
+        provider="xai",
+        context_length=131072,
+        input_cost_per_million=3.00,
+        output_cost_per_million=15.00,
+    ),
+    "x-ai/grok-3-mini": ModelInfo(
+        id="x-ai/grok-3-mini",
+        name="Grok 3 Mini",
+        provider="xai",
+        context_length=131072,
+        input_cost_per_million=0.30,
+        output_cost_per_million=0.50,
+    ),
+    # ==========================================================================
+    # DeepSeek
+    # ==========================================================================
+    "deepseek/deepseek-chat": ModelInfo(
+        id="deepseek/deepseek-chat",
+        name="DeepSeek V3",
+        provider="deepseek",
+        context_length=64000,
+        input_cost_per_million=0.14,
+        output_cost_per_million=0.28,
+    ),
+    "deepseek/deepseek-r1": ModelInfo(
+        id="deepseek/deepseek-r1",
+        name="DeepSeek R1",
+        provider="deepseek",
+        context_length=64000,
+        input_cost_per_million=0.55,
+        output_cost_per_million=2.19,
+    ),
+    "deepseek/deepseek-r1-distill-llama-70b": ModelInfo(
+        id="deepseek/deepseek-r1-distill-llama-70b",
+        name="DeepSeek R1 Distill Llama 70B",
+        provider="deepseek",
+        context_length=128000,
+        input_cost_per_million=0.23,
+        output_cost_per_million=0.69,
+    ),
+    "deepseek/deepseek-r1-distill-qwen-32b": ModelInfo(
+        id="deepseek/deepseek-r1-distill-qwen-32b",
+        name="DeepSeek R1 Distill Qwen 32B",
+        provider="deepseek",
+        context_length=128000,
+        input_cost_per_million=0.14,
+        output_cost_per_million=0.28,
+    ),
+    "deepseek/deepseek-r1:free": ModelInfo(
+        id="deepseek/deepseek-r1:free",
+        name="DeepSeek R1 (Free)",
+        provider="deepseek",
+        context_length=64000,
+        input_cost_per_million=0.0,
+        output_cost_per_million=0.0,
+    ),
+    # ==========================================================================
+    # Meta (Llama)
+    # ==========================================================================
+    "meta-llama/llama-3.3-70b-instruct": ModelInfo(
+        id="meta-llama/llama-3.3-70b-instruct",
+        name="Llama 3.3 70B",
+        provider="meta",
+        context_length=131072,
+        input_cost_per_million=0.12,
+        output_cost_per_million=0.30,
+    ),
+    "meta-llama/llama-3.3-70b-instruct:free": ModelInfo(
+        id="meta-llama/llama-3.3-70b-instruct:free",
+        name="Llama 3.3 70B (Free)",
+        provider="meta",
+        context_length=131072,
+        input_cost_per_million=0.0,
+        output_cost_per_million=0.0,
+    ),
+    "meta-llama/llama-3.1-405b-instruct": ModelInfo(
+        id="meta-llama/llama-3.1-405b-instruct",
+        name="Llama 3.1 405B",
+        provider="meta",
+        context_length=131072,
+        input_cost_per_million=2.00,
+        output_cost_per_million=2.00,
+    ),
+    "meta-llama/llama-3.1-70b-instruct": ModelInfo(
+        id="meta-llama/llama-3.1-70b-instruct",
+        name="Llama 3.1 70B",
+        provider="meta",
+        context_length=131072,
+        input_cost_per_million=0.35,
+        output_cost_per_million=0.40,
+    ),
+    "meta-llama/llama-3.1-8b-instruct": ModelInfo(
+        id="meta-llama/llama-3.1-8b-instruct",
+        name="Llama 3.1 8B",
+        provider="meta",
+        context_length=131072,
+        input_cost_per_million=0.05,
+        output_cost_per_million=0.08,
+    ),
+    "meta-llama/llama-3.1-8b-instruct:free": ModelInfo(
+        id="meta-llama/llama-3.1-8b-instruct:free",
+        name="Llama 3.1 8B (Free)",
+        provider="meta",
+        context_length=131072,
+        input_cost_per_million=0.0,
+        output_cost_per_million=0.0,
+    ),
+    "meta-llama/llama-guard-3-8b": ModelInfo(
+        id="meta-llama/llama-guard-3-8b",
+        name="Llama Guard 3 8B",
+        provider="meta",
+        context_length=8192,
+        input_cost_per_million=0.05,
+        output_cost_per_million=0.05,
+    ),
+    # ==========================================================================
+    # Mistral
+    # ==========================================================================
+    "mistralai/mistral-large-2411": ModelInfo(
+        id="mistralai/mistral-large-2411",
+        name="Mistral Large",
+        provider="mistral",
+        context_length=128000,
+        input_cost_per_million=2.00,
+        output_cost_per_million=6.00,
+    ),
+    "mistralai/mistral-medium-3": ModelInfo(
+        id="mistralai/mistral-medium-3",
+        name="Mistral Medium 3",
+        provider="mistral",
+        context_length=128000,
+        input_cost_per_million=0.40,
+        output_cost_per_million=2.00,
+    ),
+    "mistralai/mistral-small-2409": ModelInfo(
+        id="mistralai/mistral-small-2409",
+        name="Mistral Small",
+        provider="mistral",
+        context_length=32000,
+        input_cost_per_million=0.20,
+        output_cost_per_million=0.60,
+    ),
+    "mistralai/mistral-small-3.1-24b-instruct": ModelInfo(
+        id="mistralai/mistral-small-3.1-24b-instruct",
+        name="Mistral Small 3.1 24B",
+        provider="mistral",
+        context_length=96000,
+        input_cost_per_million=0.10,
+        output_cost_per_million=0.30,
+    ),
+    "mistralai/ministral-8b": ModelInfo(
+        id="mistralai/ministral-8b",
+        name="Ministral 8B",
+        provider="mistral",
+        context_length=128000,
+        input_cost_per_million=0.10,
+        output_cost_per_million=0.10,
+    ),
+    "mistralai/ministral-3b": ModelInfo(
+        id="mistralai/ministral-3b",
+        name="Ministral 3B",
+        provider="mistral",
+        context_length=128000,
+        input_cost_per_million=0.04,
+        output_cost_per_million=0.04,
+    ),
+    "mistralai/codestral-2501": ModelInfo(
+        id="mistralai/codestral-2501",
+        name="Codestral",
+        provider="mistral",
+        context_length=256000,
+        input_cost_per_million=0.30,
+        output_cost_per_million=0.90,
+    ),
+    "mistralai/codestral-mamba": ModelInfo(
+        id="mistralai/codestral-mamba",
+        name="Codestral Mamba",
+        provider="mistral",
+        context_length=256000,
+        input_cost_per_million=0.25,
+        output_cost_per_million=0.25,
+    ),
+    "mistralai/pixtral-large-2411": ModelInfo(
+        id="mistralai/pixtral-large-2411",
+        name="Pixtral Large",
+        provider="mistral",
+        context_length=128000,
+        input_cost_per_million=2.00,
+        output_cost_per_million=6.00,
+        supports_vision=True,
+    ),
+    "mistralai/pixtral-12b": ModelInfo(
+        id="mistralai/pixtral-12b",
+        name="Pixtral 12B",
+        provider="mistral",
+        context_length=128000,
+        input_cost_per_million=0.10,
+        output_cost_per_million=0.10,
+        supports_vision=True,
+    ),
+    # ==========================================================================
+    # Qwen
+    # ==========================================================================
+    "qwen/qwen-2.5-72b-instruct": ModelInfo(
+        id="qwen/qwen-2.5-72b-instruct",
+        name="Qwen 2.5 72B",
+        provider="qwen",
+        context_length=131072,
+        input_cost_per_million=0.35,
+        output_cost_per_million=0.40,
+    ),
+    "qwen/qwen-2.5-32b-instruct": ModelInfo(
+        id="qwen/qwen-2.5-32b-instruct",
+        name="Qwen 2.5 32B",
+        provider="qwen",
+        context_length=131072,
+        input_cost_per_million=0.18,
+        output_cost_per_million=0.18,
+    ),
+    "qwen/qwen-2.5-7b-instruct": ModelInfo(
+        id="qwen/qwen-2.5-7b-instruct",
+        name="Qwen 2.5 7B",
+        provider="qwen",
+        context_length=131072,
+        input_cost_per_million=0.05,
+        output_cost_per_million=0.05,
+    ),
+    "qwen/qwen-2.5-coder-32b-instruct": ModelInfo(
+        id="qwen/qwen-2.5-coder-32b-instruct",
+        name="Qwen 2.5 Coder 32B",
+        provider="qwen",
+        context_length=131072,
+        input_cost_per_million=0.18,
+        output_cost_per_million=0.18,
+    ),
+    "qwen/qwq-32b-preview": ModelInfo(
+        id="qwen/qwq-32b-preview",
+        name="QwQ 32B (Reasoning)",
+        provider="qwen",
+        context_length=32768,
+        input_cost_per_million=0.12,
+        output_cost_per_million=0.18,
+    ),
+    "qwen/qwen-2.5-72b-instruct:free": ModelInfo(
+        id="qwen/qwen-2.5-72b-instruct:free",
+        name="Qwen 2.5 72B (Free)",
+        provider="qwen",
+        context_length=131072,
+        input_cost_per_million=0.0,
+        output_cost_per_million=0.0,
+    ),
+    "qwen/qwen-2-vl-72b-instruct": ModelInfo(
+        id="qwen/qwen-2-vl-72b-instruct",
+        name="Qwen 2 VL 72B",
+        provider="qwen",
+        context_length=32768,
+        input_cost_per_million=0.40,
+        output_cost_per_million=0.40,
+        supports_vision=True,
+    ),
+    # ==========================================================================
+    # Cohere
+    # ==========================================================================
+    "cohere/command-r-plus": ModelInfo(
+        id="cohere/command-r-plus",
+        name="Command R+",
+        provider="cohere",
+        context_length=128000,
+        input_cost_per_million=2.50,
+        output_cost_per_million=10.00,
+    ),
+    "cohere/command-r": ModelInfo(
+        id="cohere/command-r",
+        name="Command R",
+        provider="cohere",
+        context_length=128000,
+        input_cost_per_million=0.15,
+        output_cost_per_million=0.60,
+    ),
+    "cohere/command-r7b-12-2024": ModelInfo(
+        id="cohere/command-r7b-12-2024",
+        name="Command R 7B",
+        provider="cohere",
+        context_length=128000,
+        input_cost_per_million=0.0375,
+        output_cost_per_million=0.15,
+    ),
+    # ==========================================================================
+    # Perplexity
+    # ==========================================================================
+    "perplexity/sonar-pro": ModelInfo(
+        id="perplexity/sonar-pro",
+        name="Sonar Pro (Online)",
+        provider="perplexity",
+        context_length=200000,
+        input_cost_per_million=3.00,
+        output_cost_per_million=15.00,
+    ),
+    "perplexity/sonar": ModelInfo(
+        id="perplexity/sonar",
+        name="Sonar (Online)",
+        provider="perplexity",
+        context_length=128000,
+        input_cost_per_million=1.00,
+        output_cost_per_million=1.00,
+    ),
+    "perplexity/sonar-reasoning": ModelInfo(
+        id="perplexity/sonar-reasoning",
+        name="Sonar Reasoning (Online)",
+        provider="perplexity",
+        context_length=128000,
+        input_cost_per_million=1.00,
+        output_cost_per_million=5.00,
+    ),
+    # ==========================================================================
+    # AI21
+    # ==========================================================================
+    "ai21/jamba-1.5-large": ModelInfo(
+        id="ai21/jamba-1.5-large",
+        name="Jamba 1.5 Large",
+        provider="ai21",
+        context_length=256000,
+        input_cost_per_million=2.00,
+        output_cost_per_million=8.00,
+    ),
+    "ai21/jamba-1.5-mini": ModelInfo(
+        id="ai21/jamba-1.5-mini",
+        name="Jamba 1.5 Mini",
+        provider="ai21",
+        context_length=256000,
+        input_cost_per_million=0.20,
+        output_cost_per_million=0.40,
+    ),
+    # ==========================================================================
+    # Amazon
+    # ==========================================================================
+    "amazon/nova-pro-v1": ModelInfo(
+        id="amazon/nova-pro-v1",
+        name="Amazon Nova Pro",
+        provider="amazon",
+        context_length=300000,
+        input_cost_per_million=0.80,
+        output_cost_per_million=3.20,
+        supports_vision=True,
+    ),
+    "amazon/nova-lite-v1": ModelInfo(
+        id="amazon/nova-lite-v1",
+        name="Amazon Nova Lite",
+        provider="amazon",
+        context_length=300000,
+        input_cost_per_million=0.06,
+        output_cost_per_million=0.24,
+        supports_vision=True,
+    ),
+    "amazon/nova-micro-v1": ModelInfo(
+        id="amazon/nova-micro-v1",
+        name="Amazon Nova Micro",
+        provider="amazon",
+        context_length=128000,
+        input_cost_per_million=0.035,
+        output_cost_per_million=0.14,
+    ),
+    # ==========================================================================
+    # Microsoft
+    # ==========================================================================
+    "microsoft/phi-4": ModelInfo(
+        id="microsoft/phi-4",
+        name="Phi-4",
+        provider="microsoft",
+        context_length=16384,
+        input_cost_per_million=0.07,
+        output_cost_per_million=0.14,
+    ),
+    "microsoft/wizardlm-2-8x22b": ModelInfo(
+        id="microsoft/wizardlm-2-8x22b",
+        name="WizardLM 2 8x22B",
+        provider="microsoft",
+        context_length=65536,
+        input_cost_per_million=0.50,
+        output_cost_per_million=0.50,
+    ),
+    # ==========================================================================
+    # Nous Research
+    # ==========================================================================
+    "nousresearch/hermes-3-llama-3.1-405b": ModelInfo(
+        id="nousresearch/hermes-3-llama-3.1-405b",
+        name="Hermes 3 405B",
+        provider="nous",
+        context_length=131072,
+        input_cost_per_million=2.00,
+        output_cost_per_million=2.00,
+    ),
+    "nousresearch/hermes-3-llama-3.1-70b": ModelInfo(
+        id="nousresearch/hermes-3-llama-3.1-70b",
+        name="Hermes 3 70B",
+        provider="nous",
+        context_length=131072,
+        input_cost_per_million=0.35,
+        output_cost_per_million=0.40,
+    ),
+    # ==========================================================================
+    # Other Notable Models
+    # ==========================================================================
+    "nvidia/llama-3.1-nemotron-70b-instruct": ModelInfo(
+        id="nvidia/llama-3.1-nemotron-70b-instruct",
+        name="Nemotron 70B",
+        provider="nvidia",
+        context_length=131072,
+        input_cost_per_million=0.35,
+        output_cost_per_million=0.40,
+    ),
+    "databricks/dbrx-instruct": ModelInfo(
+        id="databricks/dbrx-instruct",
+        name="DBRX Instruct",
+        provider="databricks",
+        context_length=32768,
+        input_cost_per_million=0.75,
+        output_cost_per_million=0.75,
+    ),
+    "inflection/inflection-3-pi": ModelInfo(
+        id="inflection/inflection-3-pi",
+        name="Inflection 3 Pi",
+        provider="inflection",
+        context_length=8192,
+        input_cost_per_million=0.80,
+        output_cost_per_million=3.20,
+    ),
+    "sao10k/l3.3-euryale-70b": ModelInfo(
+        id="sao10k/l3.3-euryale-70b",
+        name="Euryale 70B",
+        provider="sao10k",
+        context_length=131072,
+        input_cost_per_million=0.50,
+        output_cost_per_million=0.60,
+    ),
+}
+
+
+class OpenRouterProvider(BaseProvider):
+    """OpenRouter - unified API for 400+ models.
+
+    OpenRouter provides access to models from OpenAI, Anthropic, Google,
+    Meta, Mistral, and many others through a single API endpoint.
+
+    No markup on provider pricing - you pay what you'd pay directly.
+    """
+
+    provider_name = "openrouter"
+    base_url = "https://openrouter.ai/api/v1"
+
+    def __init__(self, api_key: str | None = None):
+        """Initialize OpenRouter provider.
+
+        Args:
+            api_key: OpenRouter API key. If not provided, reads from
+                     OPENROUTER_API_KEY environment variable.
+        """
+        self.api_key = api_key or os.getenv("OPENROUTER_API_KEY")
+        if not self.api_key:
+            raise ProviderError(
+                "API key required. Set OPENROUTER_API_KEY or pass api_key.",
+                provider=self.provider_name,
+            )
+
+    def _get_headers(self) -> dict[str, str]:
+        """Get request headers."""
+        return {
+            "Authorization": f"Bearer {self.api_key}",
+            "Content-Type": "application/json",
+            "HTTP-Referer": "https://sandboxy.ai",
+            "X-Title": "Sandboxy",
+        }
+
+    async def complete(
+        self,
+        model: str,
+        messages: list[dict[str, Any]],
+        temperature: float = 0.7,
+        max_tokens: int = 1024,
+        **kwargs: Any,
+    ) -> ModelResponse:
+        """Send completion request via OpenRouter."""
+        start_time = time.time()
+
+        payload = {
+            "model": model,
+            "messages": messages,
+            "temperature": temperature,
+            "max_tokens": max_tokens,
+            **kwargs,
+        }
+
+        async with httpx.AsyncClient(timeout=120.0) as client:
+            try:
+                response = await client.post(
+                    f"{self.base_url}/chat/completions",
+                    headers=self._get_headers(),
+                    json=payload,
+                )
+                response.raise_for_status()
+                data = response.json()
+            except httpx.HTTPStatusError as e:
+                error_body = e.response.text
+                raise ProviderError(
+                    f"HTTP {e.response.status_code}: {error_body}",
+                    provider=self.provider_name,
+                    model=model,
+                ) from e
+            except httpx.RequestError as e:
+                raise ProviderError(
+                    f"Request failed: {e}",
+                    provider=self.provider_name,
+                    model=model,
+                ) from e
+
+        latency_ms = int((time.time() - start_time) * 1000)
+
+        # Extract response data
+        choice = data.get("choices", [{}])[0]
+        message = choice.get("message", {})
+        usage = data.get("usage", {})
+
+        # Calculate cost if we have token counts
+        input_tokens = usage.get("prompt_tokens", 0)
+        output_tokens = usage.get("completion_tokens", 0)
+        cost = self._calculate_cost(model, input_tokens, output_tokens)
+
+        return ModelResponse(
+            content=message.get("content", ""),
+            model_id=data.get("model", model),
+            latency_ms=latency_ms,
+            input_tokens=input_tokens,
+            output_tokens=output_tokens,
+            cost_usd=cost,
+            finish_reason=choice.get("finish_reason"),
+            raw_response=data,
+        )
+
+    async def stream(
+        self,
+        model: str,
+        messages: list[dict[str, Any]],
+        temperature: float = 0.7,
+        max_tokens: int = 1024,
+        **kwargs: Any,
+    ) -> AsyncIterator[str]:
+        """Stream completion response."""
+        payload = {
+            "model": model,
+            "messages": messages,
+            "temperature": temperature,
+            "max_tokens": max_tokens,
+            "stream": True,
+            **kwargs,
+        }
+
+        async with (
+            httpx.AsyncClient(timeout=120.0) as client,
+            client.stream(
+                "POST",
+                f"{self.base_url}/chat/completions",
+                headers=self._get_headers(),
+                json=payload,
+            ) as response,
+        ):
+            response.raise_for_status()
+            async for line in response.aiter_lines():
+                if line.startswith("data: "):
+                    data = line[6:]
+                    if data == "[DONE]":
+                        break
+                    try:
+                        chunk = json.loads(data)
+                        delta = chunk.get("choices", [{}])[0].get("delta", {})
+                        content = delta.get("content", "")
+                        if content:
+                            yield content
+                    except (json.JSONDecodeError, KeyError):
+                        continue
+
+    def list_models(self) -> list[ModelInfo]:
+        """List popular models available through OpenRouter."""
+        return list(OPENROUTER_MODELS.values())
+
+    def _calculate_cost(self, model: str, input_tokens: int, output_tokens: int) -> float | None:
+        """Calculate cost in USD for a request."""
+        model_info = OPENROUTER_MODELS.get(model)
+        if not model_info or not model_info.input_cost_per_million:
+            return None
+
+        input_cost = (input_tokens / 1_000_000) * model_info.input_cost_per_million
+        output_cost = (output_tokens / 1_000_000) * model_info.output_cost_per_million
+        return round(input_cost + output_cost, 6)
+
+    async def fetch_models(self) -> list[dict[str, Any]]:
+        """Fetch full model list from OpenRouter API.
+
+        Returns live model data including pricing and availability.
+        """
+        async with httpx.AsyncClient(timeout=30.0) as client:
+            response = await client.get(
+                f"{self.base_url}/models",
+                headers=self._get_headers(),
+            )
+            response.raise_for_status()
+            data = response.json()
+            return data.get("data", [])