243 lines
7.7 KiB
JSON
243 lines
7.7 KiB
JSON
{
|
|
"providers": [
|
|
{
|
|
"key": "together",
|
|
"displayName": "Together AI",
|
|
"baseUrl": "https://api.together.xyz/v1/",
|
|
"endpoint": "chat/completions",
|
|
"websiteUrl": "https://www.together.ai/",
|
|
"isCustom": false,
|
|
"models": [
|
|
{
|
|
"modelId": "meta-llama/Llama-3.3-70B-Instruct-Turbo",
|
|
"displayName": "Llama 3.3 70B Turbo",
|
|
"provider": "together",
|
|
"description": "Fast, accurate, and cost-effective open-source model."
|
|
},
|
|
{
|
|
"modelId": "meta-llama/Llama-4-Maverick-17B-Instruct",
|
|
"displayName": "Llama 4 Maverick 17B",
|
|
"provider": "together",
|
|
"description": "Next-gen efficient architecture; outperforms older 70B models."
|
|
},
|
|
{
|
|
"modelId": "deepseek-ai/DeepSeek-V3.1",
|
|
"displayName": "DeepSeek V3.1",
|
|
"provider": "together",
|
|
"description": "Latest 671B MoE model with hybrid thinking/non-thinking modes."
|
|
}
|
|
]
|
|
},
|
|
{
|
|
"key": "mistral",
|
|
"displayName": "Mistral AI",
|
|
"baseUrl": "https://api.mistral.ai/",
|
|
"endpoint": "v1/chat/completions",
|
|
"websiteUrl": "https://mistral.ai",
|
|
"isCustom": false,
|
|
"models": [
|
|
{
|
|
"modelId": "mistral-medium-latest",
|
|
"displayName": "Mistral Medium",
|
|
"provider": "mistral",
|
|
"description": "Balanced performance and cost for a wide range of tasks."
|
|
},
|
|
{
|
|
"modelId": "mistral-large-latest",
|
|
"displayName": "Mistral Large 3",
|
|
"provider": "mistral",
|
|
"description": "Flagship model with top-tier reasoning and multilingual capabilities."
|
|
}
|
|
]
|
|
},
|
|
{
|
|
"key": "openai",
|
|
"displayName": "OpenAI",
|
|
"baseUrl": "https://api.openai.com/",
|
|
"endpoint": "v1/chat/completions",
|
|
"websiteUrl": "https://platform.openai.com/",
|
|
"isCustom": false,
|
|
"models": [
|
|
{
|
|
"modelId": "gpt-5.2",
|
|
"displayName": "GPT-5.2",
|
|
"provider": "openai",
|
|
"description": "Balanced performance with enhanced reasoning and creativity."
|
|
},
|
|
{
|
|
"modelId": "gpt-5.1-instant",
|
|
"displayName": "GPT-5.1 Instant",
|
|
"provider": "openai",
|
|
"description": "The standard high-speed efficiency model replacing older 'Nano' tiers."
|
|
}
|
|
]
|
|
},
|
|
{
|
|
"key": "anthropic",
|
|
"displayName": "Anthropic",
|
|
"baseUrl": "https://api.anthropic.com/",
|
|
"endpoint": "v1/messages",
|
|
"websiteUrl": "https://www.anthropic.com/",
|
|
"isCustom": false,
|
|
"models": [
|
|
{
|
|
"modelId": "claude-opus-4-6",
|
|
"displayName": "Claude Opus 4.6",
|
|
"provider": "anthropic",
|
|
"description": "Most intelligent model for building agents and coding with 1M context."
|
|
},
|
|
{
|
|
"modelId": "claude-sonnet-4-5",
|
|
"displayName": "Claude Sonnet 4.5",
|
|
"provider": "anthropic",
|
|
"description": "Best combination of speed and intelligence with extended thinking."
|
|
}
|
|
]
|
|
},
|
|
{
|
|
"key": "deepseek",
|
|
"displayName": "DeepSeek",
|
|
"baseUrl": "https://api.deepseek.com/",
|
|
"endpoint": "chat/completions",
|
|
"websiteUrl": "https://www.deepseek.com/",
|
|
"isCustom": false,
|
|
"models": [
|
|
{
|
|
"modelId": "deepseek-reasoner",
|
|
"displayName": "DeepSeek R1",
|
|
"provider": "deepseek",
|
|
"description": "Reasoning-focused model (Chain of Thought) for complex math/code."
|
|
},
|
|
{
|
|
"modelId": "deepseek-chat",
|
|
"displayName": "DeepSeek V3.1",
|
|
"provider": "deepseek",
|
|
"description": "Latest 671B MoE with hybrid thinking/non-thinking modes, 128K context."
|
|
}
|
|
]
|
|
},
|
|
{
|
|
"key": "gemini",
|
|
"displayName": "Google Gemini",
|
|
"baseUrl": "https://generativelanguage.googleapis.com/",
|
|
"endpoint": "v1beta/models/gemini-2.5-pro:generateContent",
|
|
"websiteUrl": "https://ai.google/",
|
|
"isCustom": false,
|
|
"models": [
|
|
{
|
|
"modelId": "gemini-2.5-pro",
|
|
"displayName": "Gemini 2.5 Pro",
|
|
"provider": "gemini",
|
|
"description": "Stable release: State-of-the-art reasoning with 1M context."
|
|
},
|
|
{
|
|
"modelId": "gemini-3-pro-preview",
|
|
"displayName": "Gemini 3 Pro",
|
|
"provider": "gemini",
|
|
"description": "Top-tier reasoning model for complex agentic workflows."
|
|
}
|
|
]
|
|
},
|
|
{
|
|
"key": "openrouter",
|
|
"displayName": "OpenRouter",
|
|
"baseUrl": "https://openrouter.ai/api/",
|
|
"endpoint": "v1/chat/completions",
|
|
"websiteUrl": "https://openrouter.ai",
|
|
"isCustom": false,
|
|
"models": []
|
|
},
|
|
{
|
|
"key": "groq",
|
|
"displayName": "Groq",
|
|
"baseUrl": "https://api.groq.com/openai/",
|
|
"endpoint": "v1/chat/completions",
|
|
"websiteUrl": "https://groq.com/",
|
|
"isCustom": false,
|
|
"models": [
|
|
{
|
|
"modelId": "meta-llama/llama-4-maverick",
|
|
"displayName": "Llama 4 Maverick",
|
|
"provider": "groq",
|
|
"description": "400B MoE powerhouse with industry-leading image and text understanding."
|
|
}
|
|
]
|
|
},
|
|
{
|
|
"key": "xai",
|
|
"displayName": "xAI Grok",
|
|
"baseUrl": "https://api.x.ai/",
|
|
"endpoint": "v1/chat/completions",
|
|
"websiteUrl": "https://x.ai",
|
|
"isCustom": false,
|
|
"models": [
|
|
{
|
|
"modelId": "grok-4-1-fast-reasoning",
|
|
"displayName": "Grok 4.1 Fast",
|
|
"provider": "xai",
|
|
"description": "Fast, flexible, and capable of reasoning."
|
|
}
|
|
]
|
|
},
|
|
{
|
|
"key": "nvidia",
|
|
"displayName": "NVIDIA NIM",
|
|
"baseUrl": "https://integrate.api.nvidia.com/",
|
|
"endpoint": "v1/chat/completions",
|
|
"websiteUrl": "https://build.nvidia.com/explore",
|
|
"isCustom": false,
|
|
"models": [
|
|
{
|
|
"modelId": "meta/llama-3.3-70b-instruct",
|
|
"displayName": "Llama 3.3 70B",
|
|
"provider": "nvidia",
|
|
"description": "Standard high-performance open model accelerated by NVIDIA."
|
|
}
|
|
]
|
|
},
|
|
{
|
|
"key": "cerebras",
|
|
"displayName": "Cerebras",
|
|
"baseUrl": "https://api.cerebras.ai/",
|
|
"endpoint": "v1/chat/completions",
|
|
"websiteUrl": "https://inference.cerebras.ai/",
|
|
"isCustom": false,
|
|
"models": [
|
|
{
|
|
"modelId": "llama-3.3-70b",
|
|
"displayName": "Llama 3.3 70B (Instant)",
|
|
"provider": "cerebras",
|
|
"description": "World's fastest inference (2000+ tokens/sec) on Wafer-Scale Engines."
|
|
},
|
|
{
|
|
"modelId": "llama-4-scout",
|
|
"displayName": "Llama 4 Scout",
|
|
"provider": "cerebras",
|
|
"description": "High-quality 17B active param model running at 2,600 tokens/sec."
|
|
}
|
|
]
|
|
},
|
|
{
|
|
"key": "huggingface",
|
|
"displayName": "Hugging Face",
|
|
"baseUrl": "https://router.huggingface.co/",
|
|
"endpoint": "v1/chat/completions",
|
|
"websiteUrl": "https://huggingface.co/settings/tokens",
|
|
"isCustom": false,
|
|
"models": [
|
|
{
|
|
"modelId": "meta-llama/Llama-3.3-70B-Instruct",
|
|
"displayName": "Llama 3.3 70B",
|
|
"provider": "huggingface",
|
|
"description": "Hosted via the Hugging Face serverless router (Free tier limits apply)."
|
|
},
|
|
{
|
|
"modelId": "Qwen/Qwen2.5-72B-Instruct",
|
|
"displayName": "Qwen 2.5 72B",
|
|
"provider": "huggingface",
|
|
"description": "High-quality open model with excellent reasoning and multilingual capabilities."
|
|
}
|
|
]
|
|
}
|
|
]
|
|
} |