Albert Jiang(@AlbertQJiang):https://t.co/TAlizrIynp A lot of grinding from Leanstral 1.5 with a tiny bit of K3 orchestration get to beat GPT-6 Astra on ArXivLean. I think #tokens is counted erroneously here but yes a 6B model does yap. Great internship work by Matéo Pirio Rossignol mentored by @roman

Albert Jiang

@AlbertQJiang

L∃∀N @MistralAI 知行合一 Nullius in verba

加入 June 2020

401 正在關注 4K 粉絲

Albert Jiang@AlbertQJiang

2026.09.11 14:23

A lot of grinding from Leanstral 1.5 with a tiny bit of K3 orchestration get to beat GPT-6 Astra on ArXivLean. I think #tokens# is counted erroneously here but yes a 6B model does yap. Great internship work by Matéo Pirio Rossignol mentored by @roman_soletskyi!