Reddit users compare Qwen3.5-VL, InternVL and Gemma 4 for uncensored image captioning
TekeshiX · reddit · 2026-08-04
The author asks which LLM/VLM is best for accurately describing what is in an image, including uncensored or NSFW content.
- Gemini no longer works for these captions because it blocks such images.
- They mention Qwen3.5-VL, InternVL, Gemma 4, and JoyCaption as candidates, but are unsure which official or ablated variants can handle the task.
- The system should run within roughly 16–48 GB of VRAM, and can be used outside ComfyUI if needed.
Related event: Developers Seek Uncensored VLMs for Explicit Image Captioning(2 posts)→
More from Models
- DeepSeek v4 Flash to be tested on a prompt ChatGPT refused for moral reasons — yacineMTB · 2026-08-04
- Kimi is praised for long-horizon coding and unusually aligned behavior — casper_hansen_ · 2026-08-04
- llama.cpp patches boost DeepSeek-V4-Flash-0731 from 3.26 to 25.91 tok/s — dyn___ · 2026-08-04
- Qwen3.8-Max ranks No. 2 in Vision Arena, 13 points behind Claude Fable 5 — arena · 2026-08-04
- Polymarket puts an 82% chance on a new Google Gemini Pro model within two weeks — Polymarket · 2026-08-04
- Claude usage meter may be overstating quota after just two turns this morning — dreamwieber · 2026-08-04