vh/dotfiles
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity

CLAUDE.md: gen/summarizer seat is now qwen3.8-27b-uncensored (gen-seat/vllm-gen)

Browse Source
This commit is contained in:
Your Name
2026-09-24 08:56:51 -07:00
parent da763fc889
commit 7166e01c2e
1 changed files with 6 additions and 5 deletions
Download Patch File Download Diff File Expand all files Collapse all files
home_root/.claude/CLAUDE.md
+6 -5
View File
@@ -487,11 +487,12 @@ note says otherwise. On other boxes, check first.)
plus paid passthroughs. Essentially-**free local compute** for low-caliber, high-volume, plus paid passthroughs. Essentially-**free local compute** for low-caliber, high-volume,
parallelizable LLM work. The **always-available summarizer / classifier endpoint** is the parallelizable LLM work. The **always-available summarizer / classifier endpoint** is the
`summarizer` (+ `summarizer-large`) / `classifier` alias — the go-to for summarization, `summarizer` (+ `summarizer-large`) / `classifier` alias — the go-to for summarization,
naming/labeling, classification, and triage. Since the 2026-07 GPU re-org it's backed by naming/labeling, classification, and triage. Since 2026-08-15 it's backed by
**`qwen3.6-35b-a3b-heretic`** (efficient MoE, ~3B active) on ana-ml2 `:8015` — the same "gen" **`qwen3.8-27b-uncensored`** (JonathanColetti Heretic-abliterated Qwen3.8-27B, quantized
seat that also serves `gen` (general chat), `gen-reasoning` (thinking variant), `image-judge`, in-house to NVFP4 W4A16 + grafted MTP head, vision-intact, 262K ctx) on ana-ml2 `:8015` —
and `qwen-image-bench`. (`granite-4.1-8b` and the old `qwen3.5-9b-fp8` vision seat are retired compose project `gen-seat` / container `vllm-gen` — the same "gen" seat that also serves
from the roster.) Treat the summarizer as a standing dependency you can reach any time. The `gen` (general chat), `gen-reasoning` (thinking variant), `image-judge`, and `qwen-image-bench`.
(Prior `qwen3.6-35b-a3b-heretic` MoE and `granite-4.1-8b` are retired from the roster.) Treat the summarizer as a standing dependency you can reach any time. The
roster drifts — query the live list at `:4000/v1/models` (~30 entries as of 2026-08-14); the roster drifts — query the live list at `:4000/v1/models` (~30 entries as of 2026-08-14); the
current tool-relevant set: current tool-relevant set:
- **Embeddings / rerankers (local, free):** `qwen3-embedding`, `qwen3-reranker`, `reranker` - **Embeddings / rerankers (local, free):** `qwen3-embedding`, `qwen3-reranker`, `reranker`