unsloth-core
Unsloth-core provides the foundational framework for accelerating Large Language Model (LLM) fine-tuning. It leverages specialized kernels and optimized gradient checkpointing to achieve up to 2x faster native inference and significant VRAM savings compared to standard Hugging Face implementations.
npx skills add majiayu000/claude-skill-registry --skill unsloth-core-cuba6112-skillfactory-cef6d0f1 --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.
--- name: unsloth-core description: Core fundamentals of Unsloth for fast LLM fine-tuning, covering FastLanguageModel setup, optimized gradient checkpointing, and native inference acceleration (triggers: unsloth, FastLanguageModel, from_pretrained, get_peft_model, for_inference, gradient checkpointing). --- ## Overview Unsloth-core provides the foundational framework for accelerating Large Language Model (LLM) fine-tuning. It leverages specialized kernels and optimized gradient checkpointing to achieve up to 2x faster native inference and significant VRAM savings compared to standard Hugging Face implementations. ## When to Use - When fine-tuning Llama-3, Mistral, or Phi models and seeking maximum speed. - When limited by VRAM and needing optimized gradient checkpointing. - When deploying models for inference and requiring low-latency responses. ## Decision Tree 1. Are you starting a new fine-tuning run? - Yes: Use `FastLanguageModel.from_pretrained()`. - No (Inference only): Load with `for_inference=True`. 2. Is your VRAM usage too high? - Yes: Set `use_gradient_checkpointing = 'unsloth'`. 3. Are you ready for generation? - Yes: Call `FastLanguageModel.for_inference(model)` to ena
- Overview
- When to Use
- Decision Tree
- Workflows
- Non-Obvious Insights
- Evidence
- Scripts
- Dependencies
- References
What does the unsloth-core skill do?
Unsloth-core provides the foundational framework for accelerating Large Language Model (LLM) fine-tuning. It leverages specialized kernels and optimized gradient checkpointing to achieve up to 2x faster native inference and significant VRAM savings compared to standard Hugging Face implementations.
How do I install it?
Run `npx skills add majiayu000/claude-skill-registry --skill unsloth-core-cuba6112-skillfactory-cef6d0f1 --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 majiayu000/claude-skill-registry, a repository with 534 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.