# Unsloth

- **Canonical URI:** https://www.vioscale.ai/software/unsloth
- **Category:** Fine Tuning Platforms
- **Homepage:** https://unsloth.ai
- **Also known as:** unsloth
- **Profile claimed by vendor:** no
- **Last updated:** 2026-08-26T19:36:34.430Z

## Vioscale score

**54.9 / 100**, confidence 40% (low). Computed 2026-09-01. _(updating — computed under an earlier model version or past its freshness window.)_

Composite of weighted, independently-sourced signals (no user reviews, no vendor payment).

| Signal | Score | Weight | Contribution | Evidence present |
|---|--:|--:|--:|:--:|
| price_level | 80 | 0.052 | 4.2 | ✓ |
| reliability | 0 | 0.073 | 0 | - |
| capabilities | 86.7 | 0.084 | 7.3 | ✓ |
| repo_stars | 92 | 0.026 | 2.4 | ✓ |
| integrations | 36.8 | 0.09355555555555557 | 3.4 | ✓ |
| dependent_projects | 0 | 0.063 | 0 | ✓ |
| dev_activity | 63.1 | 0.094 | 5.9 | ✓ |
| release_cadence | 97.8 | 0.052 | 5.1 | ✓ |
| security_posture | 0 | 0.073 | 0 | - |
| package_downloads | 0 | 0.136 | 0 | - |
| security_score | 49 | 0.042 | 2.1 | ✓ |
| pricing_transparency | 25 | 0.084 | 2.1 | ✓ |
| community_qa_activity | 0 | 0.063 | 0 | - |

## Pricing

_As of 2026-08-21, [verify at source](https://unsloth.ai/pricing). Independently observed._

Hybrid · Free tier

> Free open-source tier; Pro and Enterprise plans available

| Plan | Price | Free | Commitment |
|---|---|:--:|---|
| Free | Free | ✓ | - |
| Pro | Contact sales | - | - |
| Enterprise | Contact sales | - | - |

### Free

Open-source version with core model training capabilities

- Supports Mistral, Gemma, Llama 1/2/3 models
- 4-bit and 16-bit LoRA fine-tuning
- Single GPU support
- 2x speed improvement
- 60% VRAM reduction
- Fully local and offline operation

### Pro (Contact sales)

Enhanced performance with multi-GPU support

- 2.5x faster training
- 20% less VRAM usage
- Multi-GPU support up to 8 GPUs
- 80% VRAM reduction
- Support for any use case

### Enterprise (Contact sales)

Maximum performance with multi-node deployment and accuracy improvements

- 30x faster training
- Up to 30% accuracy improvement
- 5x faster inference
- Multi-node support
- 90% VRAM reduction
- Full training support
- Customer support included

## About

Unsloth is a free, open-source desktop app that lets users run, customize, and train AI models on their own hardware with optimizations for speed and memory usage. It supports multiple GPU setups and offers paid Pro and Enterprise tiers with enhanced performance and deployment capabilities.

_Independently observed._

## Platform & deployment

- **Platforms:** macOS, Web, Linux, Windows
- **Deployment:** Cloud / SaaS, Self-hosted

## Integrations (18)

_Independently observed._

- Claude Code
- Codex
- OpenAI API
- Hugging Face
- Google
- Model Context Protocol
- Cloudflare
- FLUX
- MiniMax-H3
- Stable Diffusion
- Qwen3.8
- DeepSeek-V4
- Gemma 2
- Gemma 4
- Mistral
- Llama 1
- Llama 2
- Llama 3

## Capabilities

_The capabilities that matter for Fine Tuning Platforms. "-" = undocumented, not absent._

| Capability | Supported |
|---|:--:|
| **Capabilities** | |
| Architecture model | Open source library local GPU |
| Peft lora and qlora parameter efficient tuning | ✓ |
| Rlhf and dpo preference alignment optimization | ✓ |
| Serverless hosting of fine tuned adapters lorax | - |
| Four bit and eight bit memory quantization | ✓ |
| Flash attention and xformers compilation | ✓ |
| Distributed multi GPU orchestration fsdp deepspeed | ✓ |
| Synthetic data generation and evaluation pipeline | - |
| Native huggingface hub push pull integration | ✓ |
| Weights and biases wandb experiment tracking | - |
| SOC2 type ii | - |
| Mit or apache permissive oss license | ✓ |
| Pricing model | Free open source |

## Release history

| Version | Date | Type |
|---|---|---|
| [v0.1.803-beta](https://github.com/unslothai/unsloth/releases/tag/v0.1.803-beta) | 2026-08-25 | stable |
| [v0.1.802-beta](https://github.com/unslothai/unsloth/releases/tag/v0.1.802-beta) | 2026-08-25 | stable |
| [v0.1.801-beta](https://github.com/unslothai/unsloth/releases/tag/v0.1.801-beta) | 2026-08-20 | stable |
| [v0.1.800-beta](https://github.com/unslothai/unsloth/releases/tag/v0.1.800-beta) | 2026-08-14 | stable |
| [v0.1.702-beta](https://github.com/unslothai/unsloth/releases/tag/v0.1.702-beta) | 2026-08-13 | stable |
| [v0.1.701-beta](https://github.com/unslothai/unsloth/releases/tag/v0.1.701-beta) | 2026-08-11 | stable |
| [v0.1.70-beta](https://github.com/unslothai/unsloth/releases/tag/v0.1.70-beta) | 2026-08-11 | stable |
| [v0.1.62-beta](https://github.com/unslothai/unsloth/releases/tag/v0.1.62-beta) | 2026-08-11 | stable |
| [v0.1.61-beta](https://github.com/unslothai/unsloth/releases/tag/v0.1.61-beta) | 2026-08-10 | stable |
| [v0.1.60-beta](https://github.com/unslothai/unsloth/releases/tag/v0.1.60-beta) | 2026-08-10 | stable |
| [v0.1.527-beta](https://github.com/unslothai/unsloth/releases/tag/v0.1.527-beta) | 2026-08-09 | stable |
| [v0.1.526-beta](https://github.com/unslothai/unsloth/releases/tag/v0.1.526-beta) | 2026-08-04 | stable |
| [v0.1.512-beta](https://github.com/unslothai/unsloth/releases/tag/v0.1.512-beta) | 2026-07-29 | stable |
| [v0.1.501-beta](https://github.com/unslothai/unsloth/releases/tag/v0.1.501-beta) | 2026-07-20 | stable |
| [v0.1.49-beta](https://github.com/unslothai/unsloth/releases/tag/v0.1.49-beta) | 2026-07-15 | stable |

_… and 5 earlier release(s)._

## Security & compliance


Known vulnerabilities: 0 (0 in the last 12 months) ([source](https://advisories.ecosyste.ms/api/v1/advisories?ecosystem=pypi&package_name=roland-sloth&per_page=100)). A count reflects scale and disclosure, not quality.

## Facts

Every value below carries its source and our confidence. Facts are re-crawled on a freshness schedule.

### integrations

| Attribute | Value | Source | Retrieved | Confidence |
|---|---|---|---|---|
| integrations.count | 18 | [link](https://unsloth.ai/pricing) | 2026-08-21 | 60% (medium) |

### pricing

| Attribute | Value | Source | Retrieved | Confidence |
|---|---|---|---|---|
| pricing.free_tier | yes | [link](https://unsloth.ai/pricing) | 2026-08-21 | 60% (medium) |
| pricing.model | freemium | [link](https://unsloth.ai/pricing) | 2026-08-21 | 60% (medium) |
| pricing.price_level | low | [link](https://unsloth.ai/pricing) | 2026-08-21 | 60% (medium) |
| pricing.transparent | no | [link](https://unsloth.ai/pricing) | 2026-08-21 | 60% (medium) |

### activity

| Attribute | Value | Source | Retrieved | Confidence |
|---|---|---|---|---|
| activity.commits_last_30d | 100 | [link](https://github.com/unslothai/unsloth/pulse) | 2026-08-26 | 65% (medium) |

### adoption

| Attribute | Value | Source | Retrieved | Confidence |
|---|---|---|---|---|
| adoption.github_stars | 74,869 | [link](https://github.com/unslothai/unsloth) | 2026-08-26 | 90% (high) |
| adoption.dependent_repos | 0 | [link](https://packages.ecosyste.ms/api/v1/packages/lookup?repository_url=https%3A%2F%2Fgithub.com%2Funslothai%2Funsloth) | 2026-08-26 | 85% (high) |

### language

| Attribute | Value | Source | Retrieved | Confidence |
|---|---|---|---|---|
| language.primary | Python | [link](https://github.com/unslothai/unsloth) | 2026-08-26 | 90% (high) |

### license

| Attribute | Value | Source | Retrieved | Confidence |
|---|---|---|---|---|
| license.spdx | Apache-2.0 | [link](https://github.com/unslothai/unsloth) | 2026-08-26 | 95% (high) |

### release

| Attribute | Value | Source | Retrieved | Confidence |
|---|---|---|---|---|
| release.cadence_days | 4 | [link](https://github.com/unslothai/unsloth/releases) | 2026-08-26 | 70% (medium) |

### security

| Attribute | Value | Source | Retrieved | Confidence |
|---|---|---|---|---|
| security.scorecard | 4.9 | [link](https://api.securityscorecards.dev/projects/github.com/unslothai/unsloth) | 2026-08-26 | 90% (high) |
| security.vulnerabilities | `{"count":0,"source":"https://advisories.ecosyste.ms/api/v1/advisories?ecosystem=pypi&package_name=roland-sloth&per_page=100","last_12m":0,"max_severity":null}` | [link](https://advisories.ecosyste.ms/api/v1/advisories?ecosystem=pypi&package_name=roland-sloth&per_page=100) | 2026-08-26 | 90% (high) |

---
*Source: Vioscale (https://www.vioscale.ai/software/unsloth). Independent, evidence-based software intelligence. Cite the canonical URI.*