9B tops 72 MTEB-style tasks; 0.6B matches gemini-embedding-2 on text retrieval
antoine_chaffin · x · 2026-10-08
Text retrieval results: on a pool of 72 domain-specific MTEB-style tasks, the 9B is the best of the evaluated models and the 0.6B reaches gemini-embedding-2 performance. On ViDoRe v3 (markdown), they score 64.7 and 61.2 respectively, with the 0.6B second only to its own 9B sibling.
Related event: Perplexity Open-Sources Multimodal Embedding Models pplx-embed-v2-late(38 posts)→
More from Models
- ChatGPT Work mode vs Codex: same quota, far more tasks done per 5-hour window — sasik520 · 2026-10-08
- Epoch's InnovationEval: AI agents still far from producing real research innovations — Afinetheorem · 2026-10-08
- 113 decision models in 3 weeks: 70 built on Qwen, sub-cent per call — jonathanmalkin · 2026-10-08
- User reports Haiku 5.5 is a major workflow upgrade in screenshot post — Sorcerer12345 · 2026-10-08
- OpenRouter launches Decision Model Rankings, with typesafeai leading all categories — gaganghotra_ · 2026-10-08
- OpenAI launches Intelligent UI: ChatGPT now answers with fully interactive interfaces — gdb · 2026-10-08