Cohere's Embed 5 Fast beats Google's EmbeddingGemma 2 on 95% of 163 benchmark datasets
krypticmouse · x · 2026-10-18
Cohere announced Embed 5 Fast, a new embedding model, and benchmarked it against Google's EmbeddingGemma 2.
- Evaluated across 163 datasets using RCP-nDCG@10
- Embed 5 Fast leads on over 95% of them
- Cohere researcher bclavie added that this is nowhere near the ceiling, teasing further improvements ahead
Related event: Cohere Says Embed 5 Fast Beats Google's EmbeddingGemma 2 on 163 Datasets(3 posts)→
More from Models
- "$400/Month Can't Buy $40k of API Forever": Team Benchmarks Open Models as Fallback — ChrisRackauckas · 2026-10-18
- OpenAI's model list surfaces aurora-red/green/blue-alpha checkpoints, tipped as Fable 5.5 answer — ChrisGPT · 2026-10-18
- Solo dev's after-hours decision model tops open-weights on JevBench, beating commercial rivals — lukaszkaiser · 2026-10-18
- DeepSeek V4.1 Flash jumps on long-run agentic tasks but regresses on constraints — teortaxesTex · 2026-10-18
- Open models outperform on routine code maintenance, but only Opus finds deep defects — ChrisRackauckas · 2026-10-18
- DeepSeek V4.1 Flash enters BuildingBench top five at 1/15 of Sonnet's cost — Lianhuiq · 2026-10-18