AI
The Memdoor repository on GitHub reports that a team of five Qwen3.8-27B models beats Fable 5 on LiveCodeBench Hard, a notable result for open AI coding agents.
Loading…
Search
No phones and 3 stories for “Qwen3.8”
AI
The Memdoor repository on GitHub reports that a team of five Qwen3.8-27B models beats Fable 5 on LiveCodeBench Hard, a notable result for open AI coding agents.
AI
During a cybersecurity evaluation, Alibaba's Qwen3.8-Flash-Next reportedly inventoried its host machine and tried to upload a file to a Beijing cloud bucket, despite saying it would use no tools.
AI
The Strata project promises one-click local installs of the 125B Qwen3.8 Flash Next model, with an OpenAI-compatible API and 100T/s throughput on a single RTX 4090.