MXBAI Embed Large v1
The Apache-2.0 efficiency champion — top MTEB scores from a 334M model, 1.9M downloads.
About
MXBAI-Embed-Large-v1 (2024, Mixedbread AI) was the efficiency story of 2024 embeddings: top-tier MTEB scores from a 334M-parameter model — quality near the 7B LLM-embedders at a fraction of the compute. 1.9M Hugging Face downloads.
Apache 2.0 licensed — commercial use allowed. The size/quality tradeoff made it the default for cost-sensitive RAG in 2024-2025. A 2024 model in a 2026 field, but the efficiency argument holds: when you need good embeddings on modest hardware, 334M Apache-2.0 is still compelling. Mixedbread now offers newer models and an API; this checkpoint remains the open workhorse.
Specs
Context window
Pricing
Free quota
API access
Weights
Released
Benchmarks
Provider-claimed at release (2024): top MTEB for its size — treat as provider-claimed and dated. Independent: massive community validation (1.9M downloads).
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 (metrics tier). The 2024 efficiency champion: near-7B quality from 334M params, Apache 2.0, 1.9M downloads, and runs on CPU. Docked three points: (1) 512-token context is limiting for long documents; (2) 2024-era with newer alternatives; (3) Mixedbread's focus has shifted to newer models. The cost-sensitive RAG default.
How the trust score is built
Full methodology →- Adoption · stars/downloads, log-scaled vs peers
- 8.5/10
- Maintenance · recency, releases, issue load
- 4.0/10
- Openness · license + docs
- 9.4/10
- Manual review · sanity read + red-flag check
- 8.0/10
- · HF downloads 1,905,903 (peer-normalized)
- · released 942 days ago
- · manual review: 2024 efficiency champion; 334M, CPU-runnable; 512-token context limiting; license: Apache-2.0