zartbot.github.io personal blog
DeepSeek-v4.1 Flash: Pushing the Limits of KV Cache Compression
A deep dive into the DeepSeek-V4.1 Flash technical report: CED, CSA2, HSI, Single-Pass mHC, Engram and FP4 KV Cache — how the KV cache was compressed to just 890 bytes per token.
- Points
- 131
- HN comments
- 10
- Submissions
- 1
- Cited by
- 0 commenters
AI & Machine LearningSoftware Engineering2026
On Hacker News
| Submitted | Title on HN | Points | Comments |
|---|---|---|---|
| Sep 17, 2026 | DeepSeek-v4.1 Flash: Pushing the Limits of KV Cache Compression by mfiguiere | 131 | 10 |
