{
 "data": [
  {
   "id": "nemotron-3-nano-omni-30b-a3b",
   "canonical_slug": "nemotron-3-nano-omni-30b-a3b",
   "name": "NVIDIA Nemotron 3 Nano Omni 30B A3B Reasoning",
   "created": 1745798400,
   "description": "33B mixture-of-experts reasoning model with 3B active parameters per token. High throughput relative to its capability class. Multimodal input.",
   "hugging_face_id": "nvidia/Nemotron-3-Nano-Omni-30B-A3B-Reasoning-BF16",
   "context_length": 131072,
   "architecture": {
    "modality": "text+image->text",
    "input_modalities": [
     "text",
     "image"
    ],
    "output_modalities": [
     "text"
    ]
   },
   "pricing": {
    "prompt": "0.00000012",
    "completion": "0.00000045"
   },
   "top_provider": {
    "context_length": 131072,
    "max_completion_tokens": 32768,
    "is_moderated": false
   },
   "supported_parameters": [
    "max_tokens",
    "temperature",
    "top_p",
    "top_k",
    "stop",
    "frequency_penalty",
    "presence_penalty",
    "repetition_penalty",
    "seed",
    "reasoning",
    "include_reasoning",
    "response_format"
   ]
  },
  {
   "id": "nemotron-3.5-content-safety",
   "canonical_slug": "nemotron-3.5-content-safety",
   "name": "NVIDIA Nemotron 3.5 Content Safety",
   "created": 1749081600,
   "description": "4.3B safety classifier from NVIDIA. Screens prompts and completions for policy violations. Designed as a guard model in front of a generation model.",
   "hugging_face_id": "nvidia/Nemotron-3.5-Content-Safety",
   "context_length": 32768,
   "architecture": {
    "modality": "text->text",
    "input_modalities": [
     "text"
    ],
    "output_modalities": [
     "text"
    ]
   },
   "pricing": {
    "prompt": "0.00000005",
    "completion": "0.00000015"
   },
   "top_provider": {
    "context_length": 32768,
    "max_completion_tokens": 4096,
    "is_moderated": false
   },
   "supported_parameters": [
    "max_tokens",
    "temperature",
    "top_p",
    "top_k",
    "stop",
    "seed",
    "response_format"
   ]
  },
  {
   "id": "nemotron-nano-12b-v2-vl",
   "canonical_slug": "nemotron-nano-12b-v2-vl",
   "name": "NVIDIA Nemotron Nano 12B v2 VL",
   "created": 1751328000,
   "description": "13.2B vision-language model from NVIDIA. Accepts images alongside text for document understanding, captioning and visual question answering.",
   "hugging_face_id": "nvidia/NVIDIA-Nemotron-Nano-12B-v2-VL-BF16",
   "context_length": 131072,
   "architecture": {
    "modality": "text+image->text",
    "input_modalities": [
     "text",
     "image"
    ],
    "output_modalities": [
     "text"
    ]
   },
   "pricing": {
    "prompt": "0.0000001",
    "completion": "0.0000003"
   },
   "top_provider": {
    "context_length": 131072,
    "max_completion_tokens": 16384,
    "is_moderated": false
   },
   "supported_parameters": [
    "max_tokens",
    "temperature",
    "top_p",
    "top_k",
    "stop",
    "frequency_penalty",
    "presence_penalty",
    "repetition_penalty",
    "seed",
    "response_format"
   ]
  },
  {
   "id": "nemotron-nano-9b-v2",
   "canonical_slug": "nemotron-nano-9b-v2",
   "name": "NVIDIA Nemotron Nano 9B v2",
   "created": 1751328000,
   "description": "Compact 8.9B dense instruct model from NVIDIA. General reasoning and instruction following at low cost. Served from EU-owned hardware in Germany.",
   "hugging_face_id": "nvidia/NVIDIA-Nemotron-Nano-9B-v2",
   "context_length": 131072,
   "architecture": {
    "modality": "text->text",
    "input_modalities": [
     "text"
    ],
    "output_modalities": [
     "text"
    ]
   },
   "pricing": {
    "prompt": "0.00000008",
    "completion": "0.00000025"
   },
   "top_provider": {
    "context_length": 131072,
    "max_completion_tokens": 16384,
    "is_moderated": false
   },
   "supported_parameters": [
    "max_tokens",
    "temperature",
    "top_p",
    "top_k",
    "stop",
    "frequency_penalty",
    "presence_penalty",
    "repetition_penalty",
    "seed",
    "logit_bias",
    "response_format"
   ]
  },
  {
   "id": "north-mini-code",
   "canonical_slug": "north-mini-code",
   "name": "Cohere North Mini Code",
   "created": 1750118400,
   "description": "30.5B Apache-2.0 code model from Cohere Labs. Code generation, completion and review across mainstream languages.",
   "hugging_face_id": "CohereLabs/North-Mini-Code-1.0",
   "context_length": 131072,
   "architecture": {
    "modality": "text->text",
    "input_modalities": [
     "text"
    ],
    "output_modalities": [
     "text"
    ]
   },
   "pricing": {
    "prompt": "0.00000015",
    "completion": "0.0000005"
   },
   "top_provider": {
    "context_length": 131072,
    "max_completion_tokens": 16384,
    "is_moderated": false
   },
   "supported_parameters": [
    "max_tokens",
    "temperature",
    "top_p",
    "top_k",
    "stop",
    "frequency_penalty",
    "presence_penalty",
    "repetition_penalty",
    "seed",
    "response_format",
    "tools",
    "tool_choice"
   ]
  },
  {
   "id": "olmo-3-32b-think",
   "canonical_slug": "olmo-3-32b-think",
   "name": "AllenAI OLMo 3 32B Think",
   "created": 1748476800,
   "description": "32.2B Apache-2.0 reasoning model from the Allen Institute. Fully open training pipeline. Emits explicit reasoning traces before answering.",
   "hugging_face_id": "allenai/Olmo-3-32B-Think",
   "context_length": 65536,
   "architecture": {
    "modality": "text->text",
    "input_modalities": [
     "text"
    ],
    "output_modalities": [
     "text"
    ]
   },
   "pricing": {
    "prompt": "0.00000015",
    "completion": "0.0000006"
   },
   "top_provider": {
    "context_length": 65536,
    "max_completion_tokens": 32768,
    "is_moderated": false
   },
   "supported_parameters": [
    "max_tokens",
    "temperature",
    "top_p",
    "top_k",
    "stop",
    "frequency_penalty",
    "presence_penalty",
    "repetition_penalty",
    "seed",
    "reasoning",
    "include_reasoning",
    "response_format"
   ]
  }
 ]
}