Technical assessment of Qwen3.8-27B ahead of Max's Arena debut
arena · x · 2026-09-02
LMArena notes that before Qwen3.8-Max-0902 landed in Arena, they had also tested Qwen3.8-27B earlier this month. Peter Gostev shared his technical assessment of this smaller open-source model from Alibaba while in SF; full video linked in the post.
More from Models
- Fable 5.1 vs Opus 5: same-prompt 3D ocean test shows rain-drop detail, at 2x the price — socialwithaayan · 2026-09-02
- Gemini learns to 'watch' video selectively: 88% fewer tokens, 66% lower cost — mfckr_eth · 2026-09-02
- Gemini 3.8 Flash full benchmarks: #16 intelligence, fastest at 305 tokens/s — ArtificialAnlys · 2026-09-02
- Full Gemini 3.8 Flash benchmark results across high, medium and low reasoning — ArtificialAnlys · 2026-09-02
- Gemini 3.8 Flash outputs 48k tokens per task on high reasoning, 30% more than predecessor — ArtificialAnlys · 2026-09-02
- Gemini 3.8 Flash outputs ~300 tokens/s, needs 2.5 min per task on high reasoning — ArtificialAnlys · 2026-09-02