vision-sft
Fine-tune vision-language models (VLMs) with supervised learning on image+text data. Use when adapting a VLM to a visual domain or task, configuring frozen-vision-tower LoRA, or debugging a VLM fine-tune that trains without learning.
Profile →npx skills add wshobson/agents --skill vision-sft --agent claude-code
Same command for any agent — swap --agent for codex, cursor, copilot.
Weekly change comes from our own snapshots, not the repository page — it measures attention, not adoption.
# Vision-Language SFT This skill assumes `finetuning-method-selection` already routed here: the data shape is image+text demonstrations, not preference pairs or a verifiable reward signal, and the base is a vision-language model rather than a text-only one. `lora-qlora-recipes` covers the text-only LoRA/QLoRA recipe this skill specializes for the vision tower and projector; read that skill first if the LoRA fundamentals (rank, alpha, target modules) aren't already familiar. **Input:** an image+text dataset and a VLM base model already picked from the model catalog. **Output format:** a validated adapter config — which components are frozen, LoRA target modules, and a `min_pixels`/`max_pixels` budget — that `llm-finetuning-training-engineer` consumes directly when it generates a runnable script. ## Quick Reference | Situation | Default | |---|---| | Adapting behavior on familiar images | Frozen tower+projector, LoRA r=8–16, α=16–32 | | Visual domain shift | Unfreeze last-6 ViT layers, vision LR 5–10x lower | | Doesn't fit in bf16 at target rank | QLoRA — frozen vision tower only | | `fast_inference=True` | `finetune_vision_layers=False` | | Loss normal, eval not improving | Check th
- Quick Reference
- The Consensus Recipe
- When to Unfreeze
- The Two Silent Killers
- Unsloth Specifics
- Model Choice
- Failure Modes
- References
What does the vision-sft skill do?
Fine-tune vision-language models (VLMs) with supervised learning on image+text data. Use when adapting a VLM to a visual domain or task, configuring frozen-vision-tower LoRA, or debugging a VLM fine-tune that trains without learning.
How do I install it?
Run `npx skills add wshobson/agents --skill vision-sft --agent claude-code` — it drops the skill into your project so the agent can pick it up. Swap the --agent value for codex, cursor or copilot if you use one of those.
Where does this skill come from?
From wshobson/agents, a repository with 38,479 stars. We read it straight from the repository tree rather than a submitted listing, so what you see here is what is actually published.
Is a popular skill a good skill?
Not necessarily. Stars measure attention, not adoption — a repository can trend for a week and be abandoned. That is why we show the weekly change from our own snapshots next to the total, instead of a single flattering number.