24 lines
1.4 KiB
Markdown
24 lines
1.4 KiB
Markdown
# lfm-vl-uncensored-seat
|
|
|
|
The **abliterated LFM2.5-VL-3B** (`SC117/LFM2.5-VL-3B-Uncensored-GGUF`, Q4_K_M plus
|
|
its own mmproj Q8_0) on **nh3-ml1** `:8032`, direct only. It runs **in parallel**
|
|
with the stock seat (`stacks/lfm-vl-seat`, `:8030`) so brokkr can A/B them for
|
|
NSFW captioning. The stock model sanitized nudity and refused one image. The
|
|
operator wants an uncensored one (relayed by brokkr-smithy-dev, 2026-09-26).
|
|
|
|
- endpoint: `http://10.100.50.80:8032/v1/chat/completions`, model `lfm25-vl-3b-uncensored`
|
|
- same llama.cpp build (b11176) and flags as the stock seat. The mmproj is SC117's
|
|
own export, which differs slightly from LiquidAI's (different sha256) and is
|
|
used as shipped.
|
|
- positive control 2026-09-26: the synthetic shapes and text image was read
|
|
exactly, same as stock. Decode ~102 tok/s.
|
|
- "Uncensored" is the repo's claim (abliterix Trial 65). brokkr's 6-image eval
|
|
decides whether it engages and stays coherent. Fallbacks, if it fails, are in
|
|
brokkr's request: the Heretic builds by MuXodious or AX1Y2JP, or an in-house
|
|
Heretic pass.
|
|
- ✅ **Passed brokkr's 6-image eval, 2026-09-26 0058.** It engages where stock
|
|
hard-refused or sanitized, and it stays coherent (no catatonia or word salad).
|
|
The control image was read exactly, and with no image it answers "empty". Its
|
|
limit is fine detail (a 3B model), not censorship. brokkr: keep the seat up, no
|
|
fallback needed. It stays direct-only; brokkr has not asked for a gateway alias.
|