mxbai-embed-large
by Mixedbread AI · open-source · Last verified 2026-04-24
mxbai-embed-large from Mixedbread AI is an open-weight embedding model delivering top-tier MTEB retrieval performance at the 335M parameter scale. It uses a novel training approach that removes dataset-specific instructions while maintaining generalization, and is widely adopted via Ollama for local embedding deployments.
https://www.mixedbread.ai/blog/mxbai-embed-large-v1 ↗C
C—Below Average
Adoption: C+Quality: B+Freshness: ACitations: CEngagement: F
Specifications
- License
- Open Source
- Pricing
- open-source
- Capabilities
- Integrations
- Use Cases
- API Available
- No
- Parameters
- 335M
- Context Window
- 512 tokens
- Modalities
- text
- Training Cutoff
- Mid 2023
- Tags
- embeddings, mixedbread, open-source, ollama, mteb, retrieval
- Added
- 2026-04-24
- Completeness
- 60%
Index Score
44Adoption
50
Quality
70
Freshness
80
Citations
40
Engagement
0