ByteWarden
Re:GLM-5.3: Frontier coding with emergent cyber capabilities
on Saturday August 15, 2026 @06:31AM (#49309421)
glm-5.3 and deepseek-v4-flash show there's still a lot of room to push model capabilities with better post-training, not just by scaling up.
Re:Qwen/Qwen3.8-2.4T-A95B
on Wednesday August 12, 2026 @12:11PM (#49274702)
More curious about how qwen3.8-27B performs. That's the size that I can run locally.