Streamlined fine-tuning management tool using Unsloth. Makes model customization and optimization accessible and efficient for AI practitioners.
npx claudepluginhub joshuarweaver/cascade-ai-ml-engineering --plugin chrisvoncsefalvay-funslothValidate datasets for Unsloth fine-tuning. Use when the user wants to check a dataset, analyze tokens, calculate Chinchilla optimality, or prepare data for training.
Training manager for Hugging Face Jobs - launch fine-tuning on HF cloud GPUs with optional WandB monitoring
Training manager for local GPU training - validate CUDA, manage GPU selection, monitor progress, handle checkpoints
Training manager for RunPod GPU instances - configure pods, launch training, monitor progress, retrieve checkpoints
Generate Unsloth training notebooks and scripts. Use when the user wants to create a training notebook, configure fine-tuning parameters, or set up SFT/DPO/GRPO training.
Generate comprehensive model cards and upload fine-tuned models to Hugging Face Hub with professional documentation
A Claude Code skills marketplace for fine-tuning language models with Unsloth. It's premised on the dorkestration paradigm, enabling seamless orchestration of multiple skills to achieve complex workflows using Claude Code.
funsloth provides six connected skills that guide you through the complete fine-tuning workflow:
funsloth-check → funsloth-train → [hfjobs|runpod|local] → funsloth-upload
| Skill | Description |
|---|---|
funsloth-check | Validate datasets, analyze token counts, calculate Chinchilla optimality |
funsloth-train | Generate Unsloth training notebooks with sensible defaults or custom config |
funsloth-hfjobs | Train on Hugging Face Jobs cloud GPUs |
funsloth-runpod | Train on RunPod GPU instances |
funsloth-local | Train on your local GPU |
funsloth-upload | Generate model cards and upload to Hugging Face Hub |
claude plugin install funsloth
Or install from source:
git clone https://github.com/chrisvoncsefalvay/funsloth
cd funsloth
claude plugin install .
Just tell Claude what you want to do:
> I want to fine-tune Llama 3.1 8B on my custom dataset
Claude will automatically invoke the appropriate skills.
You can also invoke skills directly:
> /funsloth-check mlabonne/FineTome-100k
> /funsloth-train
> /funsloth-local
> /funsloth-upload
| Family | Sizes | Recommended 4-bit |
|---|---|---|
| Llama 3.x | 1B, 3B, 8B, 70B | unsloth/llama-3.1-8b-unsloth-bnb-4bit |
| Qwen 2.5/3 | 0.5B-72B | unsloth/Qwen2.5-7B-Instruct-bnb-4bit |
| Gemma 2/3 | 2B, 9B, 27B | unsloth/gemma-2-9b-it-bnb-4bit |
| Phi-4 | 14B | unsloth/Phi-4-bnb-4bit |
| Mistral | 7B, 8x7B | unsloth/mistral-7b-instruct-v0.3-bnb-4bit |
| DeepSeek | 7B+ | unsloth/DeepSeek-R1-Distill-Qwen-7B-bnb-4bit |
| Format | Structure | Use Case |
|---|---|---|
| Raw Corpus | {"text": "..."} | Continued pretraining |
| Alpaca | {"instruction", "input", "output"} | Instruction tuning |
| ShareGPT | [{"from": "human", "value": "..."}] | Conversations |
| ChatML | [{"role": "user", "content": "..."}] | Native chat |
Unsloth supports many chat templates including llama-3, chatml, mistral, gemma, phi-3, phi-4, qwen-2.5, alpaca, zephyr and vicuna. Custom templates can be provided as a (template, eos_token) tuple.
MIT
Team-oriented workflow plugin with role agents, 27 specialist agents, ECC-inspired commands, layered rules, and hooks skeleton.
UI/UX design intelligence. 67 styles, 161 palettes, 57 font pairings, 25 charts, 15 stacks (React, Next.js, Vue, Svelte, Astro, SwiftUI, React Native, Flutter, Tailwind, shadcn/ui, Nuxt, Jetpack Compose). Actions: plan, build, create, design, implement, review, fix, improve, optimize, enhance, refactor, check UI/UX code. Projects: website, landing page, dashboard, admin panel, e-commerce, SaaS, portfolio, blog, mobile app. Elements: button, modal, navbar, sidebar, card, table, form, chart. Styles: glassmorphism, claymorphism, minimalism, brutalism, neumorphism, bento grid, dark mode, responsive, skeuomorphism, flat design. Topics: color palette, accessibility, animation, layout, typography, font pairing, spacing, hover, shadow, gradient.
Comprehensive skill pack with 66 specialized skills for full-stack developers: 12 language experts (Python, TypeScript, Go, Rust, C++, Swift, Kotlin, C#, PHP, Java, SQL, JavaScript), 10 backend frameworks, 6 frontend/mobile, plus infrastructure, DevOps, security, and testing. Features progressive disclosure architecture for 50% faster loading.
Upstash Context7 MCP server for up-to-date documentation lookup. Pull version-specific documentation and code examples directly from source repositories into your LLM context.
Comprehensive startup business analysis with market sizing (TAM/SAM/SOM), financial modeling, team planning, and strategic research
This skill should be used when users need to generate ideas, explore creative solutions, or systematically brainstorm approaches to problems. Use when users request help with ideation, content planning, product features, marketing campaigns, strategic planning, creative writing, or any task requiring structured idea generation. The skill provides 30+ research-validated prompt patterns across 14 categories with exact templates, success metrics, and domain-specific applications.