Open-Source 2B Embedding Model topk-embed-v1-small Hits Frontier Ranking on Perplexity Benchmark
antoine_chaffin · x · 2026-10-02
Marek Galovic announced that their open-source topk-embed-v1-small (2B) embedding model achieves frontier-level ranking performance (combined nDCG@10) on Perplexity's private Q2D-Web benchmark, with the eval run by Antoine Chaffin's team. A notable result for a small open embedding model competing with frontier retrieval systems.
More from Models
- Grok joins X group chats, can generate images and videos on requester's quota — mark_k · 2026-10-03
- State of Local AI 2026: one gaming GPU now matches the world's best model from Feb 2026 — Scobleizer · 2026-10-03
- Leak: DeepSeek V4.1 Pro ~2T params, trained sparse from scratch with no instabilities — teortaxesTex · 2026-10-03
- Report: DeepSeek views Kimi's approach as hard to scale, seeks a cheap scalable recipe — teortaxesTex · 2026-10-03
- Running 180B Qwen3.8-Flash-Next at 40-50 t/s on 64GB RAM + 16GB VRAM with Strata — danamir_ · 2026-10-03
- AI slide design is even more conspicuous than AI writing — siglesias · 2026-10-03