readme: add plain-english collapse at top

2026-09-01 10:23:36 +10:00
parent 5ddea3477f
commit eb6e00759c
1 changed files with 21 additions and 0 deletions
@@ -16,6 +16,27 @@
 </div>
 <details>
 <summary><b>plain english — what was done</b></summary>
 qwen3.8-flash-next is the new qwen model that is great for coding — it beats
 claude opus 4.6 on swe-bench and runs on a $2500 mini pc.
 1. the official conversion code missed a step — every hyper-connection norm
   was off by exactly 1.0, so the first quant printed garbage. we found it,
   fixed the converter, and added a test so it never happens again.
 2. we made the model smaller without losing quality — the big 51b n-gram table
   tolerates 4-bit, saving 27g — then proved it across context depths from
   0 to 256k.
 3. we kept everything that makes strix halo fast — vulkan kernels, graph
   reuse, speculative decoding with the 4b draft head.
 if you just want to run it: `docker compose up --build` and open
 `http://localhost:8080`. pick the 91g file for speed, the 116g file if you
 need 262k context.
 </details>
 ---
 ## results