73 lines
1.9 KiB
JSON
73 lines
1.9 KiB
JSON
{
|
|
"chat_models": [
|
|
{
|
|
"id": "llama-3.3-70b-versatile",
|
|
"context": 131072,
|
|
"speed": "280 T/sec",
|
|
"description": "Llama 3.3 70B versatile model with 131K context"
|
|
},
|
|
{
|
|
"id": "llama-3.1-8b-instant",
|
|
"context": 131072,
|
|
"speed": "560 T/sec",
|
|
"description": "Fast Llama 3.1 8B instant model with 131K context"
|
|
},
|
|
{
|
|
"id": "openai/gpt-oss-120b",
|
|
"context": 131072,
|
|
"speed": "500 T/sec",
|
|
"description": "OpenAI GPT-OSS 120B model with 131K context"
|
|
},
|
|
{
|
|
"id": "openai/gpt-oss-20b",
|
|
"context": 131072,
|
|
"speed": "1000 T/sec",
|
|
"description": "OpenAI GPT-OSS 20B model with 131K context"
|
|
},
|
|
{
|
|
"id": "groq/compound",
|
|
"context": 131072,
|
|
"speed": "450 T/sec",
|
|
"description": "Agentic model with web search, code execution, and reasoning"
|
|
},
|
|
{
|
|
"id": "groq/compound-mini",
|
|
"context": 131072,
|
|
"speed": "450 T/sec",
|
|
"description": "Lighter agentic model with web search and reasoning capabilities"
|
|
},
|
|
{
|
|
"id": "meta-llama/llama-guard-4-12b",
|
|
"context": 131072,
|
|
"speed": "1200 T/sec",
|
|
"description": "Content screening and safety model"
|
|
}
|
|
],
|
|
"vision_models": [
|
|
{
|
|
"id": "meta-llama/llama-4-scout-17b-16e-instruct",
|
|
"context": 131072,
|
|
"max_image_size": "20MB",
|
|
"description": "Llama 4 Scout 17B multimodal model"
|
|
},
|
|
{
|
|
"id": "meta-llama/llama-4-maverick-17b-128e-instruct",
|
|
"context": 131072,
|
|
"max_image_size": "20MB",
|
|
"description": "Llama 4 Maverick 17B multimodal model"
|
|
}
|
|
],
|
|
"audio_models": [
|
|
{
|
|
"id": "whisper-large-v3",
|
|
"context": 448,
|
|
"description": "Whisper large v3 for audio transcription"
|
|
},
|
|
{
|
|
"id": "whisper-large-v3-turbo",
|
|
"context": 448,
|
|
"description": "Faster Whisper large v3 turbo variant"
|
|
}
|
|
]
|
|
}
|