← All free tools · 🧩 Open Source
Unsloth
✓ Verified against the official pageFast LLM fine-tuning with lower memory use — on a single consumer GPU.
Free limit
Free limit
Open code: self-host free
Access type
Open source
Period
—
Watermark
Not verified
Commercial use
Unclear
Credit card
Not verified
Sign-up
Not verified
API
Yes: Указано: "Connect: Serve models through an OpenAI compatible API" — локальный OpenAI-совместимый API для запуска моделей
Resolution and duration limits
—
Regional restrictions
Not specified
Checked
Nov 22, 2026
Unsloth is an open-source project (Apache-2.0, AGPL-3.0): the desktop app, web UI and code are freely available, plus free notebooks for training models. No numeric quotas, limits or payment/card requirements are stated on the page.
What the tool does
LoRA/QLoRA, faster training, GGUF export.
Official source
Train for free with our notebooks.
- github.com/unslothai/unsloth — Nov 22, 2026 (18899 симв.)
Free-tier change history
| Field | Before | After | Date |
|---|---|---|---|
| No changes recorded yet. | |||