BLIP-3
Salesforce's open multimodal family (xGen-MM) — vision-language models you can actually download.
About
BLIP-3, also published as xGen-MM, is Salesforce AI Research's family of open large multimodal models from 2024. The largest variant has 4 billion parameters and handles interleaved image-text input for captioning, VQA, and document understanding. Salesforce open-sourced the weights, training data recipes, and fine-tuning code — though licensing varies by checkpoint, with some under Apache 2.0 and others under CC-BY-NC-4.0, so check the specific card before commercial use.
Specs
Context window
Pricing
Free quota
API access
Weights
Released
Benchmarks
Paper-reported: competitive with similar-size open LMMs on standard VQA and captioning benchmarks.
Provider-reported unless a source says otherwise. Benchmarks are marketing-adjacent by nature — treat them as directional.
Use cases
Trust score — 7/10
Scored 7/10 (verified tier, metrics-v1). Evidence: xGen-MM paper Aug 2024; weights on HF under Salesforce org; License varies by checkpoint: Apache-2.0 vs CC-BY-NC-4.0; Check the specific HF card before commercial use. License: Mixed: some checkpoints Apache-2.0, others CC-BY-NC-4.0 — verify per checkpoint on the HF card..
How the trust score is built
Full methodology →- Adoption · stars/downloads, log-scaled vs peers
- 7.0/10
- Maintenance · recency, releases, issue load
- 7.0/10
- Openness · license + docs
- 7.0/10
- Manual review · sanity read + red-flag check
- 7.0/10
- · xGen-MM paper Aug 2024; weights on HF under Salesforce org
- · License varies by checkpoint: Apache-2.0 vs CC-BY-NC-4.0
- · Check the specific HF card before commercial use