KV cache quantization hurts DeepSeek V4 Flash quality
Quantizing DeepSeek V4 Flash's KV cache from BF16 to Q8 strongly degrades perplexity, KL divergence, and Same-TopP metrics — a striking contrast to Qwen 397B.
2 sources
Daily brief
Get tomorrow's AI brief in your inbox
More stories today
- Z.ai CEO Jie Tang: GLM 5.3 gains come from RL, not parameter count
- New tool adds 14 skills to Claude Code and Cursor for Markdown diagrams
- Tool turns Claude into a team of AI employees on your Mac
- GOP panics over Big Tech ties as Trump shifts on AI regulation
- Ethan Mollick: Claude's skill creator beats ChatGPT for reusable skills