dank/serverAppConfigs
1
0
Fork 0
Code Issues 6 Pull Requests Actions Packages Projects Releases Wiki Activity

docs(frank-medium): record the Qwen-Sharp v22.4.0 chat template

Browse Source 
Runbook deployed-flags table and AGENTS.md now name the template
(froggeric v22.4.0, repo-tracked, terse by default) and the rollback
target; the stale qwen-fixed-v22.3.jinja row and the long-gone
MM_MERGE_ENGINE note are gone.
This commit is contained in:
Frank
2026-09-21 11:18:58 -07:00
parent 6bccff3cd2
commit 3b28b2afb7
2 changed files with 5 additions and 3 deletions
Download Patch File Download Diff File Expand all files Collapse all files
AGENTS.md
+3 -1
View File
@@ -348,7 +348,9 @@ full window; the DFlash2 drafter's 2.1 GiB costs the pool vs MTP's 461,540).
DFlash2 was A/B tested against MTP self-speculation on this checkpoint and
wins on decode (100.0 vs 85.9 t/s) — see the runbook. Tools via `qwen3_coder`,
reasoning via `qwen3` (the field is `reasoning`, not llama.cpp's
`reasoning_content`), vision through the Gemma4 merge path.
`reasoning_content`), vision through the Gemma4 merge path, chat template
Qwen-Sharp v22.4.0 (froggeric, repo-tracked at `dankserver/frank-medium/`,
terse system prompt by default).
**The entrypoint is quantization-aware.** `radiance-entrypoint.sh` reads
`quantization_config.quant_method` from the checkpoint's `config.json` and
serverAppConfigs-docs/LLM/frank-medium-radiance-mxfp4.md
+2 -2
View File
@@ -47,7 +47,7 @@ board. The collective that actually runs is vLLM's R4D one-shot allreduce, which
does not use RCCL at all (TheRock RCCL has no gfx1201 device code; see
[[Dual-R9700-Cooperation]]).
## Deployed flags (2026-09-20)
## Deployed flags (2026-09-21)
From `dankserver/frank-medium/quadlet/frank-medium.container`:
@@ -64,7 +64,7 @@ From `dankserver/frank-medium/quadlet/frank-medium.container`:
| Parsers | `--enable-auto-tool-choice --tool-call-parser qwen3_coder`, `--reasoning-parser qwen3` |
| Sampler floor | `--override-sampling-effects` (temp 0.7 / top_p 0.95 / top_k 20) |
| Multimodal | `--limit-mm-per-prompt '{"image": 8}'` |
| Chat template | `qwen-fixed-v22.3.jinja` (patch dir bind-mounted, `MM_MERGE_ENGINE=AXROW_SEPARATE` for Gemma4 vision) |
| Chat template | `qwen-sharp-v22.4.0.jinja` (froggeric Qwen-Sharp v22.4.0, repo-tracked at `dankserver/frank-medium/`, mounted at `/templates/`; terse system prompt by default — opt out per-request with `chat_template_kwargs {"terse": false}`; rollback: flip `CHAT_TEMPLATE` back to `/patches/qwen-fixed-v22.3.jinja`) |
**The KV pool is auto-sized, not pinned.** Production ships no `KV_MEM` pin
(`KV_MEM=0`): the pool comes up at whatever VRAM the weights + drafter leave.