Qwen3.8-27B Goes Live on Cerebras at Full Speed, Scoring 34 on AA Index
Scobleizer · x · 2026-09-12
Cerebras has launched Alibaba's open-weight dense model Qwen3.8-27B on its inference stack at Cerebras speed. The model scores 34 on the Artificial Analysis Intelligence Index, putting it in the same band as GPT-5.6 Luna, DeepSeek V4 Pro, and Claude Sonnet 4.6.
More from Infra
- Sandbox latency compounds into a hidden infra tax in coding-agent RL, Daytona intern finds — tw_killian · 2026-09-12
- Unsloth UD-quants for coding: is 8-bit worth it over faster 6-bit? — Jorlen · 2026-09-12
- OpenAI details how Python-based storage platform Habitat scaled to 1 billion ChatGPT users — TheMoonMidas · 2026-09-12
- Intel's silicon photonics couplers hit 1-1.5 dB IL, with visible epoxy delamination flaws — jwt0625 · 2026-09-12
- Alibaba's Taotian S266 encoder sweeps 6 world titles at MSU 2025 video codec contest — 机器之心 · 2026-09-12
- CPO paper criticized for vague DLW-to-PIC coupling description: 'such as TCB' — jwt0625 · 2026-09-12