hermes/haloq38flash
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

readme: tagline under title — what was done, not just what it is

Browse Source
This commit is contained in:
Julian Beltran
2026-09-01 00:07:42 +10:00
parent 2555344204
commit 7b9c8151b3
1 changed files with 3 additions and 1 deletions
Download Patch File Download Diff File Expand all files Collapse all files
README.md
+3 -1
View File
@@ -2,7 +2,9 @@
# haloq38flash
**qwen3.8-flash-next on amd strix halo — 91g quant, 56 tok/s, 262k context**
**qwen3.8-flash-next (125b-a6b) on amd strix halo — 56 tok/s mtp, 262k context, 91g provenance-verified quant**
*converter bug found + fixed · vulkan fa/mmq kernel tuning · greedy-oracle validated speculative decoding · 51b n-gram table cut to 4 bits · ssd streaming to 262k*
[![HF Model](https://img.shields.io/badge/🤗_Model-IQ4_XS_PLE-ffD21E)](https://huggingface.co/julianmb/Qwen3.8-Flash-Next-IQ4_XS-GGUF)
[![Speed](https://img.shields.io/badge/MTP_@8k-56.4_t%2Fs-brightgreen)](#results)