OpenSearch's first 600M dense embedding model beats its Qwen3 base on MTEB
tomaarsen · x · 2026-11-13
OpenSearch's first dense embedding model, opensearch-neural-embed-v1, scores 73.76 on MTEB(eng, v2) Mean (Task), beating its Qwen3-Embedding-0.6B base at 70.47, with 58.37 average nDCG@10 across MTEB's 15 BEIR tasks. Per the official card, the 598M model ranks #1 among sub-1B single-vector dense models on BEIR, sits within 1.5 points of Qwen3-Embedding-8B at 13x smaller, and was trained label-free by distilling two teachers on 17M unlabeled English texts.
Related event: OpenSearch Releases First Open-Source Embedding and Sparse Retrieval Models(3 posts)→
More from Models
- Cloudflare hosts Thinking Machines' inkling 2 small on Workers AI at day-0 speed — michellechen · 2026-11-14
- Masked decoding trick tames text rivers in Gemma 3 12B, with a width-aware LoRA and coherence RL — cephaloform · 2026-11-14
- 250,000-human AI mafia game finds OpenAI models lie most, Anthropic snitches — Scobleizer · 2026-11-14
- Developer downgrades Claude Ultra to Pro and moves all work to Codex over 5-hour rate limits — umeshonai · 2026-11-14
- Pangram details per-domain false positive rates (1 in 10,000), enabling profitable AI detection — soldni · 2026-11-14
- Blogger predicts GPT-6.2 launches before Nov 19, floats three reasons for the delay — kimmonismus · 2026-11-14