AllenAI: Molmo2 8B

Description

Molmo2-8B is an open vision-language model developed by the Allen Institute for AI (Ai2) as part of the Molmo2 family, supporting image, video, and multi-image understanding and grounding. It is based on Qwen3-8B and uses SigLIP 2 as its vision backbone, outperforming other open-weight, open-data models on short videos, counting, and captioning, while remaining competitive on long-video tasks.

Raw fields snapshot

{
  "id": "allenai/molmo-2-8b",
  "name": "AllenAI: Molmo2 8B",
  "description": "Molmo2-8B is an open vision-language model developed by the Allen Institute for AI (Ai2) as part of the Molmo2 family, supporting image, video, and multi-image understanding and grounding. It is based on Qwen3-8B and uses SigLIP 2 as its vision backbone, outperforming other open-weight, open-data models on short videos, counting, and captioning, while remaining competitive on long-video tasks.",
  "created": 1767996672,
  "canonical_slug": "allenai/molmo-2-8b-20260109",
  "hugging_face_id": "allenai/Molmo2-8B",
  "source_type": "openrouter_only",
  "context_length": 36864,
  "max_completion_tokens": 36864,
  "is_moderated": false,
  "architecture": {
    "modality": "text+image+video->text",
    "input_modalities": [
      "text",
      "image",
      "video"
    ],
    "output_modalities": [
      "text"
    ],
    "tokenizer": "Other",
    "instruct_type": null
  },
  "input_modalities": [
    "text",
    "image",
    "video"
  ],
  "output_modalities": [
    "text"
  ],
  "modality": "text+image+video->text",
  "tokenizer": "Other",
  "instruct_type": null,
  "supported_parameters": [
    "frequency_penalty",
    "logit_bias",
    "max_tokens",
    "presence_penalty",
    "repetition_penalty",
    "seed",
    "stop",
    "temperature",
    "top_k",
    "top_p"
  ],
  "default_parameters": {
    "temperature": null,
    "top_p": null,
    "frequency_penalty": null
  },
  "per_request_limits": null,
  "top_provider": {
    "context_length": 36864,
    "max_completion_tokens": 36864,
    "is_moderated": false
  },
  "pricing": {
    "prompt": "0.0000002",
    "completion": "0.0000002"
  },
  "PPM": {
    "prompt": 0.2,
    "completion": 0.2
  },
  "openrouter_raw": {
    "id": "allenai/molmo-2-8b",
    "canonical_slug": "allenai/molmo-2-8b-20260109",
    "hugging_face_id": "allenai/Molmo2-8B",
    "name": "AllenAI: Molmo2 8B",
    "created": 1767996672,
    "description": "Molmo2-8B is an open vision-language model developed by the Allen Institute for AI (Ai2) as part of the Molmo2 family, supporting image, video, and multi-image understanding and grounding. It is based on Qwen3-8B and uses SigLIP 2 as its vision backbone, outperforming other open-weight, open-data models on short videos, counting, and captioning, while remaining competitive on long-video tasks.",
    "context_length": 36864,
    "architecture": {
      "modality": "text+image+video->text",
      "input_modalities": [
        "text",
        "image",
        "video"
      ],
      "output_modalities": [
        "text"
      ],
      "tokenizer": "Other",
      "instruct_type": null
    },
    "pricing": {
      "prompt": "0.0000002",
      "completion": "0.0000002"
    },
    "top_provider": {
      "context_length": 36864,
      "max_completion_tokens": 36864,
      "is_moderated": false
    },
    "per_request_limits": null,
    "supported_parameters": [
      "frequency_penalty",
      "logit_bias",
      "max_tokens",
      "presence_penalty",
      "repetition_penalty",
      "seed",
      "stop",
      "temperature",
      "top_k",
      "top_p"
    ],
    "default_parameters": {
      "temperature": null,
      "top_p": null,
      "frequency_penalty": null
    },
    "expiration_date": null
  }
}