Molmo2-8B

Name: Molmo2-8B
Author: allenai

by allenai

Open VLM with video grounding: locate and track objects across frames

85Kdl/month

8Bparams

HuggingFace Run it on your own data, free

Identifiers

Model ID

allenai/Molmo2-8B

Feature URI

mixpeek://image_extractor@v1/allenai_molmo2_8b_v1

Overview

Molmo2 is a fully open (weights + data) vision-language model from AI2 that supports image, video, and multi-image understanding with strong spatial grounding. It can point to, track, and count objects in video, outperforming Qwen3-VL on video counting (35.5 vs 29.6) and Gemini 3 Pro on video pointing (38.4 vs 20.0 F1).

Built on Qwen3-8B and SigLIP 2 vision encoder, Molmo2 is unique in offering both open weights and open training data, enabling full reproducibility.

Architecture

8B parameter VLM using Qwen3-8B language backbone + SigLIP 2 vision encoder. Multi-image and video input via frame sampling. Spatial grounding via coordinate prediction in output tokens.

Mixpeek SDK Integration

// No extractor parameter takes a Hugging Face model id (checked against
// GET /v1/discovery/extractors, which returns 13), so Molmo2-8B runs
// on your side and the output is upserted through POST
// /v1/namespaces/{namespace_id}/documents/upsert. On Enterprise the other
// path is to upload the weights instead: POST /v1/namespaces/{id}/models
// accepts the huggingface format and a custom plugin loads them.
const res = await fetch(
  "https://api.mixpeek.com/v1/namespaces/ns_your_namespace/documents/upsert",
  {
    method: "POST",
    headers: {
      Authorization: "Bearer API_KEY",
      "Content-Type": "application/json",
    },
    body: JSON.stringify({
      collection_id: "col_your_collection",
      documents: [
        {
          document_id: "asset-00412",
          // The model produces text, so it lands in payload. Give the
          // collection a text vector index and embed that text to make it
          // searchable rather than only filterable.
          payload: { extracted_text: modelOutput, source_key: "archive/2026/asset-00412" },
          vectors: { "multimodal-embedding": embeddingOfModelOutput },
        },
      ],
    }),
  },
);

// Managed alternative, if this exact model is not the requirement:
// universal_extractor@v1 runs google/gemini-embedding-2
// (3072-d) over a bucket, with no inference of your own.