All payments in the preview are in test mode. Read more

Code & Dev Tools · verified 2026-07-19

Unsloth

Fine-tune open models on one GPU.

Optimised LoRA and QLoRA training that cuts memory use enough to fine-tune useful models on a single consumer card.

Rating: 4.5 / 5
Pricing: Open source
Best for: Fine-tuning, Self-hosting, Cost control
Platforms: Linux, Windows, CLI
Runs locally: Yes
Content policy: Standard filters

Matching filters

Writing code Research Video editing Fine-tuning & training Model hosting & inference Persistent memory Open source Free or open source Free when self-hosted Windows Linux Desktop app CLI / terminal Runs locally

What stands out

2x faster LoRA training
Free Colab notebooks
Exports to GGUF

Also in Code & Dev Tools

Cursor

The editor that knows your repo.

Claude Code

An agent that lives in your terminal.

GitHub Copilot

Autocomplete that stayed and grew up.