npm - @lobehub/chat - Versions diffs - 1.16.7 → 1.16.8 - Mend

@lobehub/chat 1.16.7 → 1.16.8

This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.

Potentially problematic release.

This version of @lobehub/chat might be problematic. Click here for more details.

Files changed (20) hide show

package/CHANGELOG.md +25 -0
package/README.md +8 -8
package/README.zh-CN.md +8 -8
package/package.json +1 -1
package/src/config/modelProviders/ai360.ts +23 -68
package/src/config/modelProviders/anthropic.ts +39 -2
package/src/config/modelProviders/baichuan.ts +30 -11
package/src/config/modelProviders/bedrock.ts +80 -19
package/src/config/modelProviders/deepseek.ts +11 -2
package/src/config/modelProviders/fireworksai.ts +6 -2
package/src/config/modelProviders/google.ts +53 -2
package/src/config/modelProviders/groq.ts +46 -7
package/src/config/modelProviders/minimax.ts +2 -0
package/src/config/modelProviders/openai.ts +102 -1
package/src/const/discover.ts +1 -0
package/src/libs/agent-runtime/openai/__snapshots__/index.test.ts.snap +56 -0
package/src/server/routers/edge/config/__snapshots__/index.test.ts.snap +32 -0
package/src/server/routers/edge/config/index.test.ts +4 -0
package/src/types/llm.ts +29 -1
package/src/utils/parseModels.test.ts +8 -0

package/src/config/modelProviders/bedrock.ts CHANGED Viewed

@@ -1,12 +1,11 @@
 import { ModelProviderCard } from '@/types/llm';
-// ref https://docs.aws.amazon.com/bedrock/latest/userguide/model-ids.html
 // ref https://docs.aws.amazon.com/bedrock/latest/userguide/conversation-inference.html
 // ref https://us-east-1.console.aws.amazon.com/bedrock/home?region=us-east-1#/models
 // ref https://us-west-2.console.aws.amazon.com/bedrock/home?region=us-west-2#/models
 const Bedrock: ModelProviderCard = {
   chatModels: [
-/*
+    /*
     // TODO: Not support for now
     {
       description: 'Amazon Titan Text Lite is a light weight efficient model ideal for fine-tuning for English-language tasks, including like summarization and copywriting, where customers want a smaller, more cost-effective model that is also highly customizable.',
@@ -28,96 +27,156 @@ const Bedrock: ModelProviderCard = {
     },
 */
     {
-      description: 'Claude 3.5 Sonnet raises the industry bar for intelligence, outperforming competitor models and Claude 3 Opus on a wide range of evaluations, with the speed and cost of our mid-tier model, Claude 3 Sonnet.',
+      description:
+        'Claude 3.5 Sonnet raises the industry bar for intelligence, outperforming competitor models and Claude 3 Opus on a wide range of evaluations, with the speed and cost of our mid-tier model, Claude 3 Sonnet.',
       displayName: 'Claude 3.5 Sonnet',
       enabled: true,
       functionCall: true,
       id: 'anthropic.claude-3-5-sonnet-20240620-v1:0',
+      pricing: {
+        input: 3,
+        output: 15,
+      },
       tokens: 200_000,
       vision: true,
     },
     {
-      description: 'Claude 3 Sonnet by Anthropic strikes the ideal balance between intelligence and speed—particularly for enterprise workloads. It offers maximum utility at a lower price than competitors, and is engineered to be the dependable, high-endurance workhorse for scaled AI deployments. Claude 3 Sonnet can process images and return text outputs, and features a 200K context window.',
+      description:
+        'Claude 3 Sonnet by Anthropic strikes the ideal balance between intelligence and speed—particularly for enterprise workloads. It offers maximum utility at a lower price than competitors, and is engineered to be the dependable, high-endurance workhorse for scaled AI deployments. Claude 3 Sonnet can process images and return text outputs, and features a 200K context window.',
       displayName: 'Claude 3 Sonnet',
       enabled: true,
       functionCall: true,
       id: 'anthropic.claude-3-sonnet-20240229-v1:0',
+      pricing: {
+        input: 3,
+        output: 15,
+      },
       tokens: 200_000,
       vision: true,
     },
     {
-      description: 'Claude 3 Opus is Anthropic most powerful AI model, with state-of-the-art performance on highly complex tasks. It can navigate open-ended prompts and sight-unseen scenarios with remarkable fluency and human-like understanding. Claude 3 Opus shows us the frontier of what’s possible with generative AI. Claude 3 Opus can process images and return text outputs, and features a 200K context window.',
+      description:
+        'Claude 3 Opus is Anthropic most powerful AI model, with state-of-the-art performance on highly complex tasks. It can navigate open-ended prompts and sight-unseen scenarios with remarkable fluency and human-like understanding. Claude 3 Opus shows us the frontier of what’s possible with generative AI. Claude 3 Opus can process images and return text outputs, and features a 200K context window.',
       displayName: 'Claude 3 Opus',
       enabled: true,
       functionCall: true,
       id: 'anthropic.claude-3-opus-20240229-v1:0',
+      pricing: {
+        input: 15,
+        output: 75,
+      },
       tokens: 200_000,
       vision: true,
     },
     {
-      description: 'Claude 3 Haiku is Anthropic fastest, most compact model for near-instant responsiveness. It answers simple queries and requests with speed. Customers will be able to build seamless AI experiences that mimic human interactions. Claude 3 Haiku can process images and return text outputs, and features a 200K context window.',
+      description:
+        'Claude 3 Haiku is Anthropic fastest, most compact model for near-instant responsiveness. It answers simple queries and requests with speed. Customers will be able to build seamless AI experiences that mimic human interactions. Claude 3 Haiku can process images and return text outputs, and features a 200K context window.',
       displayName: 'Claude 3 Haiku',
       enabled: true,
       functionCall: true,
       id: 'anthropic.claude-3-haiku-20240307-v1:0',
+      pricing: {
+        input: 0.25,
+        output: 1.25,
+      },
       tokens: 200_000,
       vision: true,
     },
     {
-      description: 'An update to Claude 2 that features double the context window, plus improvements across reliability, hallucination rates, and evidence-based accuracy in long document and RAG contexts.',
+      description:
+        'An update to Claude 2 that features double the context window, plus improvements across reliability, hallucination rates, and evidence-based accuracy in long document and RAG contexts.',
       displayName: 'Claude 2.1',
       id: 'anthropic.claude-v2:1',
+      pricing: {
+        input: 8,
+        output: 24,
+      },
       tokens: 200_000,
     },
     {
-      description: 'Anthropic highly capable model across a wide range of tasks from sophisticated dialogue and creative content generation to detailed instruction following.',
+      description:
+        'Anthropic highly capable model across a wide range of tasks from sophisticated dialogue and creative content generation to detailed instruction following.',
       displayName: 'Claude 2.0',
       id: 'anthropic.claude-v2',
+      pricing: {
+        input: 8,
+        output: 24,
+      },
       tokens: 100_000,
     },
     {
-      description: 'A fast, affordable yet still very capable model, which can handle a range of tasks including casual dialogue, text analysis, summarization, and document question-answering.',
+      description:
+        'A fast, affordable yet still very capable model, which can handle a range of tasks including casual dialogue, text analysis, summarization, and document question-answering.',
       displayName: 'Claude Instant',
       id: 'anthropic.claude-instant-v1',
+      pricing: {
+        input: 0.8,
+        output: 2.4,
+      },
       tokens: 100_000,
     },
     {
-      description: 'An update to Meta Llama 3 8B Instruct that includes an expanded 128K context length, multilinguality and improved reasoning capabilities. The Llama 3.1 offering of multilingual large language models (LLMs) is a collection of pretrained and instruction-tuned generative models in 8B, 70B and 405B sizes (text in/text out). The Llama 3.1 instruction-tuned text only models (8B, 70B, 405B) are optimized for multilingual dialogue use cases and outperform many of the available open source chat models on common industry benchmarks. Llama 3.1 is intended for commercial and research use in multiple languages. Instruction tuned text only models are intended for assistant-like chat, whereas pretrained models can be adapted for a variety of natural language generation tasks. The Llama 3.1 models also support the ability to leverage the outputs of its models to improve other models including synthetic data generation and distillation. Llama 3.1 is an auto-regressive language model that uses an optimized transformer architecture. The tuned versions use supervised fine-tuning (SFT) and reinforcement learning with human feedback (RLHF) to align with human preferences for helpfulness and safety.',
+      description:
+        'An update to Meta Llama 3 8B Instruct that includes an expanded 128K context length, multilinguality and improved reasoning capabilities. The Llama 3.1 offering of multilingual large language models (LLMs) is a collection of pretrained and instruction-tuned generative models in 8B, 70B and 405B sizes (text in/text out). The Llama 3.1 instruction-tuned text only models (8B, 70B, 405B) are optimized for multilingual dialogue use cases and outperform many of the available open source chat models on common industry benchmarks. Llama 3.1 is intended for commercial and research use in multiple languages. Instruction tuned text only models are intended for assistant-like chat, whereas pretrained models can be adapted for a variety of natural language generation tasks. The Llama 3.1 models also support the ability to leverage the outputs of its models to improve other models including synthetic data generation and distillation. Llama 3.1 is an auto-regressive language model that uses an optimized transformer architecture. The tuned versions use supervised fine-tuning (SFT) and reinforcement learning with human feedback (RLHF) to align with human preferences for helpfulness and safety.',
       displayName: 'Llama 3.1 8B Instruct',
       enabled: true,
       functionCall: true,
       id: 'meta.llama3-1-8b-instruct-v1:0',
+      pricing: {
+        input: 0.22,
+        output: 0.22,
+      },
       tokens: 128_000,
     },
     {
-      description: 'An update to Meta Llama 3 70B Instruct that includes an expanded 128K context length, multilinguality and improved reasoning capabilities. The Llama 3.1 offering of multilingual large language models (LLMs) is a collection of pretrained and instruction-tuned generative models in 8B, 70B and 405B sizes (text in/text out). The Llama 3.1 instruction-tuned text only models (8B, 70B, 405B) are optimized for multilingual dialogue use cases and outperform many of the available open source chat models on common industry benchmarks. Llama 3.1 is intended for commercial and research use in multiple languages. Instruction tuned text only models are intended for assistant-like chat, whereas pretrained models can be adapted for a variety of natural language generation tasks. The Llama 3.1 models also support the ability to leverage the outputs of its models to improve other models including synthetic data generation and distillation. Llama 3.1 is an auto-regressive language model that uses an optimized transformer architecture. The tuned versions use supervised fine-tuning (SFT) and reinforcement learning with human feedback (RLHF) to align with human preferences for helpfulness and safety.',
+      description:
+        'An update to Meta Llama 3 70B Instruct that includes an expanded 128K context length, multilinguality and improved reasoning capabilities. The Llama 3.1 offering of multilingual large language models (LLMs) is a collection of pretrained and instruction-tuned generative models in 8B, 70B and 405B sizes (text in/text out). The Llama 3.1 instruction-tuned text only models (8B, 70B, 405B) are optimized for multilingual dialogue use cases and outperform many of the available open source chat models on common industry benchmarks. Llama 3.1 is intended for commercial and research use in multiple languages. Instruction tuned text only models are intended for assistant-like chat, whereas pretrained models can be adapted for a variety of natural language generation tasks. The Llama 3.1 models also support the ability to leverage the outputs of its models to improve other models including synthetic data generation and distillation. Llama 3.1 is an auto-regressive language model that uses an optimized transformer architecture. The tuned versions use supervised fine-tuning (SFT) and reinforcement learning with human feedback (RLHF) to align with human preferences for helpfulness and safety.',
       displayName: 'Llama 3.1 70B Instruct',
       enabled: true,
       functionCall: true,
       id: 'meta.llama3-1-70b-instruct-v1:0',
+      pricing: {
+        input: 0.99,
+        output: 0.99,
+      },
       tokens: 128_000,
     },
     {
-      description: 'Meta Llama 3.1 405B Instruct is the largest and most powerful of the Llama 3.1 Instruct models that is a highly advanced model for conversational inference and reasoning, synthetic data generation, and a base to do specialized continual pre-training or fine-tuning on a specific domain. The Llama 3.1 offering of multilingual large language models (LLMs) is a collection of pretrained and instruction-tuned generative models in 8B, 70B and 405B sizes (text in/text out). The Llama 3.1 instruction-tuned text only models (8B, 70B, 405B) are optimized for multilingual dialogue use cases and outperform many of the available open source chat models on common industry benchmarks. Llama 3.1 is intended for commercial and research use in multiple languages. Instruction tuned text only models are intended for assistant-like chat, whereas pretrained models can be adapted for a variety of natural language generation tasks. The Llama 3.1 models also support the ability to leverage the outputs of its models to improve other models including synthetic data generation and distillation. Llama 3.1 is an auto-regressive language model that uses an optimized transformer architecture. The tuned versions use supervised fine-tuning (SFT) and reinforcement learning with human feedback (RLHF) to align with human preferences for helpfulness and safety.',
+      description:
+        'Meta Llama 3.1 405B Instruct is the largest and most powerful of the Llama 3.1 Instruct models that is a highly advanced model for conversational inference and reasoning, synthetic data generation, and a base to do specialized continual pre-training or fine-tuning on a specific domain. The Llama 3.1 offering of multilingual large language models (LLMs) is a collection of pretrained and instruction-tuned generative models in 8B, 70B and 405B sizes (text in/text out). The Llama 3.1 instruction-tuned text only models (8B, 70B, 405B) are optimized for multilingual dialogue use cases and outperform many of the available open source chat models on common industry benchmarks. Llama 3.1 is intended for commercial and research use in multiple languages. Instruction tuned text only models are intended for assistant-like chat, whereas pretrained models can be adapted for a variety of natural language generation tasks. The Llama 3.1 models also support the ability to leverage the outputs of its models to improve other models including synthetic data generation and distillation. Llama 3.1 is an auto-regressive language model that uses an optimized transformer architecture. The tuned versions use supervised fine-tuning (SFT) and reinforcement learning with human feedback (RLHF) to align with human preferences for helpfulness and safety.',
       displayName: 'Llama 3.1 405B Instruct',
       enabled: true,
       functionCall: true,
       id: 'meta.llama3-1-405b-instruct-v1:0',
+      pricing: {
+        input: 5.32,
+        output: 16,
+      },
       tokens: 128_000,
     },
     {
-      description: 'Meta Llama 3 is an accessible, open large language model (LLM) designed for developers, researchers, and businesses to build, experiment, and responsibly scale their generative AI ideas. Part of a foundational system, it serves as a bedrock for innovation in the global community. Ideal for limited computational power and resources, edge devices, and faster training times.',
+      description:
+        'Meta Llama 3 is an accessible, open large language model (LLM) designed for developers, researchers, and businesses to build, experiment, and responsibly scale their generative AI ideas. Part of a foundational system, it serves as a bedrock for innovation in the global community. Ideal for limited computational power and resources, edge devices, and faster training times.',
       displayName: 'Llama 3 8B Instruct',
       id: 'meta.llama3-8b-instruct-v1:0',
+      pricing: {
+        input: 0.3,
+        output: 0.6,
+      },
       tokens: 8000,
     },
     {
-      description: 'Meta Llama 3 is an accessible, open large language model (LLM) designed for developers, researchers, and businesses to build, experiment, and responsibly scale their generative AI ideas. Part of a foundational system, it serves as a bedrock for innovation in the global community. Ideal for content creation, conversational AI, language understanding, R&D, and Enterprise applications.',
+      description:
+        'Meta Llama 3 is an accessible, open large language model (LLM) designed for developers, researchers, and businesses to build, experiment, and responsibly scale their generative AI ideas. Part of a foundational system, it serves as a bedrock for innovation in the global community. Ideal for content creation, conversational AI, language understanding, R&D, and Enterprise applications.',
       displayName: 'Llama 3 70B Instruct',
       id: 'meta.llama3-70b-instruct-v1:0',
+      pricing: {
+        input: 2.65,
+        output: 3.5,
+      },
       tokens: 8000,
     },
-/*
+    /*
     // TODO: Not support for now
     {
       description: 'A 7B dense Transformer, fast-deployed and easily customisable. Small, yet powerful for a variety of use cases. Supports English and code, and a 32k context window.',
@@ -157,7 +216,7 @@ const Bedrock: ModelProviderCard = {
       tokens: 32_000,
     },
 */
-/*
+    /*
     // TODO: Not support for now
     {
       description: 'Command R+ is a highly performant generative language model optimized for large scale production workloads.',
@@ -176,7 +235,7 @@ const Bedrock: ModelProviderCard = {
       tokens: 128_000,
     },
 */
-/*
+    /*
     // Cohere Command (Text) and AI21 Labs Jurassic-2 (Text) don't support chat with the Converse API
     {
       description: 'Command is Cohere flagship text generation model. It is trained to follow user commands and to be instantly useful in practical business applications.',
@@ -191,7 +250,7 @@ const Bedrock: ModelProviderCard = {
       tokens: 4000,
     },
 */
-/*
+    /*
     // TODO: Not support for now
     {
       description: 'The latest Foundation Model from AI21 Labs, Jamba-Instruct offers an impressive 256K context window and delivers the best value per price on core text generation, summarization, and question answering tasks for the enterprise.',
@@ -200,7 +259,7 @@ const Bedrock: ModelProviderCard = {
       tokens: 256_000,
     },
 */
-/*
+    /*
     // Cohere Command (Text) and AI21 Labs Jurassic-2 (Text) don't support chat with the Converse API
     {
       description: 'Jurassic-2 Mid is less powerful than Ultra, yet carefully designed to strike the right balance between exceptional quality and affordability. Jurassic-2 Mid can be applied to any language comprehension or generation task including question answering, summarization, long-form copy generation, advanced information extraction and many others.',
@@ -218,7 +277,9 @@ const Bedrock: ModelProviderCard = {
   ],
   checkModel: 'anthropic.claude-instant-v1',
   id: 'bedrock',
+  modelsUrl: 'https://docs.aws.amazon.com/bedrock/latest/userguide/model-ids.html',
   name: 'Bedrock',
+  url: 'https://docs.aws.amazon.com/bedrock/latest/userguide/what-is-bedrock.html',
 };
 export default Bedrock;

package/src/config/modelProviders/deepseek.ts CHANGED Viewed

@@ -4,18 +4,27 @@ import { ModelProviderCard } from '@/types/llm';
 const DeepSeek: ModelProviderCard = {
   chatModels: [
     {
-      description: '擅长通用对话任务',
-      displayName: 'DeepSeek V2.5',
+      description:
+        '融合通用与代码能力的全新开源模型, 不仅保留了原有 Chat 模型的通用对话能力和 Coder 模型的强大代码处理能力，还更好地对齐了人类偏好。此外，DeepSeek-V2.5 在写作任务、指令跟随等多个方面也实现了大幅提升。',
+      displayName: 'DeepSeek-V2.5',
       enabled: true,
       functionCall: true,
       id: 'deepseek-chat',
+      pricing: {
+        cachedInput: 0.014,
+        input: 0.14,
+        output: 0.28,
+      },
+      releasedAt: '2024-09-05',
       tokens: 128_000,
     },
   ],
   checkModel: 'deepseek-chat',
   id: 'deepseek',
   modelList: { showModelFetcher: true },
+  modelsUrl: 'https://platform.deepseek.com/api-docs/zh-cn/quick_start/pricing',
   name: 'DeepSeek',
+  url: 'https://deepseek.com',
 };
 export default DeepSeek;

package/src/config/modelProviders/fireworksai.ts CHANGED Viewed

@@ -5,7 +5,8 @@ import { ModelProviderCard } from '@/types/llm';
 const FireworksAI: ModelProviderCard = {
   chatModels: [
     {
-      description: 'Fireworks latest and most performant function-calling model. Firefunction-v2 is based on Llama-3 and trained to excel at function-calling as well as chat and instruction-following. See blog post for more details https://fireworks.ai/blog/firefunction-v2-launch-post',
+      description:
+        'Fireworks latest and most performant function-calling model. Firefunction-v2 is based on Llama-3 and trained to excel at function-calling as well as chat and instruction-following. See blog post for more details https://fireworks.ai/blog/firefunction-v2-launch-post',
       displayName: 'Firefunction V2',
       enabled: true,
       functionCall: true,
@@ -20,7 +21,8 @@ const FireworksAI: ModelProviderCard = {
       tokens: 32_768,
     },
     {
-      description: 'Vision-language model allowing both image and text as inputs (single image is recommended), trained on OSS model generated training data and open sourced on huggingface at fireworks-ai/FireLLaVA-13b',
+      description:
+        'Vision-language model allowing both image and text as inputs (single image is recommended), trained on OSS model generated training data and open sourced on huggingface at fireworks-ai/FireLLaVA-13b',
       displayName: 'FireLLaVA-13B',
       enabled: true,
       functionCall: false,
@@ -137,7 +139,9 @@ const FireworksAI: ModelProviderCard = {
   checkModel: 'accounts/fireworks/models/firefunction-v2',
   id: 'fireworksai',
   modelList: { showModelFetcher: true },
+  modelsUrl: 'https://fireworks.ai/models?show=Serverless',
   name: 'Fireworks AI',
+  url: 'https://fireworks.ai',
 };
 export default FireworksAI;

package/src/config/modelProviders/google.ts CHANGED Viewed

@@ -1,7 +1,5 @@
 import { ModelProviderCard } from '@/types/llm';
-// ref https://ai.google.dev/models/gemini
-// api https://ai.google.dev/api/rest/v1beta/models/list
 const Google: ModelProviderCard = {
   chatModels: [
     {
@@ -11,6 +9,11 @@ const Google: ModelProviderCard = {
       functionCall: true,
       id: 'gemini-1.5-flash-latest',
       maxOutput: 8192,
+      pricing: {
+        cachedInput: 0.018_75,
+        input: 0.075,
+        output: 0.3,
+      },
       tokens: 1_048_576 + 8192,
       vision: true,
     },
@@ -21,6 +24,12 @@ const Google: ModelProviderCard = {
       functionCall: true,
       id: 'gemini-1.5-flash-exp-0827',
       maxOutput: 8192,
+      pricing: {
+        cachedInput: 0.018_75,
+        input: 0.075,
+        output: 0.3,
+      },
+      releasedAt: '2024-08-27',
       tokens: 1_048_576 + 8192,
       vision: true,
     },
@@ -30,6 +39,7 @@ const Google: ModelProviderCard = {
       functionCall: true,
       id: 'gemini-1.5-flash-8b-exp-0827',
       maxOutput: 8192,
+      releasedAt: '2024-08-27',
       tokens: 1_048_576 + 8192,
       vision: true,
     },
@@ -49,6 +59,12 @@ const Google: ModelProviderCard = {
       functionCall: true,
       id: 'gemini-1.5-pro-latest',
       maxOutput: 8192,
+      pricing: {
+        cachedInput: 0.875,
+        input: 3.5,
+        output: 10.5,
+      },
+      releasedAt: '2024-02-15',
       tokens: 2_097_152 + 8192,
       vision: true,
     },
@@ -59,6 +75,12 @@ const Google: ModelProviderCard = {
       functionCall: true,
       id: 'gemini-1.5-pro-exp-0827',
       maxOutput: 8192,
+      pricing: {
+        cachedInput: 0.875,
+        input: 3.5,
+        output: 10.5,
+      },
+      releasedAt: '2024-08-27',
       tokens: 2_097_152 + 8192,
       vision: true,
     },
@@ -68,6 +90,12 @@ const Google: ModelProviderCard = {
       functionCall: true,
       id: 'gemini-1.5-pro-exp-0801',
       maxOutput: 8192,
+      pricing: {
+        cachedInput: 0.875,
+        input: 3.5,
+        output: 10.5,
+      },
+      releasedAt: '2024-08-01',
       tokens: 2_097_152 + 8192,
       vision: true,
     },
@@ -77,6 +105,12 @@ const Google: ModelProviderCard = {
       functionCall: true,
       id: 'gemini-1.5-pro-001',
       maxOutput: 8192,
+      pricing: {
+        cachedInput: 0.875,
+        input: 3.5,
+        output: 10.5,
+      },
+      releasedAt: '2024-02-15',
       tokens: 2_097_152 + 8192,
       vision: true,
     },
@@ -86,6 +120,11 @@ const Google: ModelProviderCard = {
       displayName: 'Gemini 1.0 Pro',
       id: 'gemini-1.0-pro-latest',
       maxOutput: 2048,
+      pricing: {
+        input: 0.5,
+        output: 1.5,
+      },
+      releasedAt: '2023-12-06',
       tokens: 30_720 + 2048,
     },
     {
@@ -95,6 +134,11 @@ const Google: ModelProviderCard = {
       functionCall: true,
       id: 'gemini-1.0-pro-001',
       maxOutput: 2048,
+      pricing: {
+        input: 0.5,
+        output: 1.5,
+      },
+      releasedAt: '2023-12-06',
       tokens: 30_720 + 2048,
     },
     {
@@ -103,11 +147,17 @@ const Google: ModelProviderCard = {
       displayName: 'Gemini 1.0 Pro 002 (Tuning)',
       id: 'gemini-1.0-pro-002',
       maxOutput: 2048,
+      pricing: {
+        input: 0.5,
+        output: 1.5,
+      },
+      releasedAt: '2023-12-06',
       tokens: 30_720 + 2048,
     },
   ],
   checkModel: 'gemini-1.5-flash-latest',
   id: 'google',
+  modelsUrl: 'https://ai.google.dev/gemini-api/docs/models/gemini',
   name: 'Google',
   proxyUrl: {
     placeholder: 'https://generativelanguage.googleapis.com',
@@ -116,6 +166,7 @@ const Google: ModelProviderCard = {
     speed: 2,
     text: true,
   },
+  url: 'https://ai.google.dev',
 };
 export default Google;

package/src/config/modelProviders/groq.ts CHANGED Viewed

@@ -1,6 +1,5 @@
 import { ModelProviderCard } from '@/types/llm';
-// ref https://console.groq.com/docs/models
 // ref https://console.groq.com/docs/tool-use
 const Groq: ModelProviderCard = {
   chatModels: [
@@ -10,6 +9,11 @@ const Groq: ModelProviderCard = {
       enabled: true,
       functionCall: true,
       id: 'llama-3.1-8b-instant',
+      maxOutput: 8192,
+      pricing: {
+        input: 0.05,
+        output: 0.08,
+      },
       tokens: 131_072,
     },
     {
@@ -17,9 +21,14 @@ const Groq: ModelProviderCard = {
       enabled: true,
       functionCall: true,
       id: 'llama-3.1-70b-versatile',
+      maxOutput: 8192,
+      pricing: {
+        input: 0.59,
+        output: 0.79,
+      },
       tokens: 131_072,
     },
-/*
+    /*
     // Offline due to overwhelming demand! Stay tuned for updates.
     {
       displayName: 'Llama 3.1 405B (Preview)',
@@ -29,17 +38,25 @@ const Groq: ModelProviderCard = {
     },
 */
     {
-      displayName: 'Llama 3 Groq 8B Tool Use (Preview)',
+      displayName: 'Llama 3 Groq 8B Tool Use Preview 8K',
       enabled: true,
       functionCall: true,
       id: 'llama3-groq-8b-8192-tool-use-preview',
+      pricing: {
+        input: 0.19,
+        output: 0.19,
+      },
       tokens: 8192,
     },
     {
-      displayName: 'Llama 3 Groq 70B Tool Use (Preview)',
+      displayName: 'Llama 3 Groq 70B Tool Use Preview 8K',
       enabled: true,
       functionCall: true,
       id: 'llama3-groq-70b-8192-tool-use-preview',
+      pricing: {
+        input: 0.89,
+        output: 0.89,
+      },
       tokens: 8192,
     },
     {
@@ -47,6 +64,10 @@ const Groq: ModelProviderCard = {
       enabled: true,
       functionCall: true,
       id: 'llama3-8b-8192',
+      pricing: {
+        input: 0.05,
+        output: 0.08,
+      },
       tokens: 8192,
     },
     {
@@ -54,26 +75,42 @@ const Groq: ModelProviderCard = {
       enabled: true,
       functionCall: true,
       id: 'llama3-70b-8192',
+      pricing: {
+        input: 0.59,
+        output: 0.79,
+      },
       tokens: 8192,
     },
     {
-      displayName: 'Gemma 2 9B',
+      displayName: 'Gemma 2 9B 8k',
       enabled: true,
       functionCall: true,
       id: 'gemma2-9b-it',
+      pricing: {
+        input: 0.2,
+        output: 0.2,
+      },
       tokens: 8192,
     },
     {
-      displayName: 'Gemma 7B',
+      displayName: 'Gemma 7B 8k Instruct',
       functionCall: true,
       id: 'gemma-7b-it',
+      pricing: {
+        input: 0.07,
+        output: 0.07,
+      },
       tokens: 8192,
     },
     {
-      displayName: 'Mixtral 8x7B',
+      displayName: 'Mixtral 8x7B Instruct 32k',
       enabled: true,
       functionCall: true,
       id: 'mixtral-8x7b-32768',
+      pricing: {
+        input: 0.24,
+        output: 0.24,
+      },
       tokens: 32_768,
     },
     {
@@ -86,10 +123,12 @@ const Groq: ModelProviderCard = {
   ],
   checkModel: 'gemma2-9b-it',
   id: 'groq',
+  modelsUrl: 'https://console.groq.com/docs/models',
   name: 'Groq',
   proxyUrl: {
     placeholder: 'https://api.groq.com/openai/v1',
   },
+  url: 'https://groq.com',
 };
 export default Groq;

package/src/config/modelProviders/minimax.ts CHANGED Viewed

@@ -42,11 +42,13 @@ const Minimax: ModelProviderCard = {
   ],
   checkModel: 'abab6.5s-chat',
   id: 'minimax',
+  modelsUrl: 'https://platform.minimaxi.com/document/Models',
   name: 'Minimax',
   smoothing: {
     speed: 2,
     text: true,
   },
+  url: 'https://www.minimaxi.com',
 };
 export default Minimax;