/jump to repo

Repos Recent Bookmarks Watched Notes Tags Discover Compare Stats GitHub

Repository brief

unslothai/unsloth

Read the upstream summary on the left, browse the cached forks below it, and load each fork comparison into the right-hand panel.

Cached analysis

cached 2026-03-30T15:43:43.368Z

1mo ago

unslothai/unsloth

Unsloth is an active Python open source repository for a local AI training and inference stack. The repo is heavily starred and forked, is not archived, and its README describes Unsloth Studio as a web UI for training and running open models locally across Windows, Linux, and macOS.

Loading tags...

Stars58,657

Forks4,971

Default branchmain

Last pushed2026-03-30T13:40:49Z

Recommended shortcuts

Jump straight into Discofork's strongest cached fork picks, or open a compare view in one click.

Forks

Choose a fork to inspect

10 of 10 fork briefs

Maintenance:

Magnitude:

Sort:

Selected

Choose upstream unless you have a concrete reason to stay on an old, divergent fork. This fork looks like a stale 2024-era snapshot with substantial divergence, so it is better suited to niche compatibility or legacy workflows than to ongoing use.

Choose this fork only if you specifically want the older QLoRA-focused behavior and can live without the modern Studio and inference stack. For most adopters, upstream is the safer choice because it is active and materially more complete.

Choose this fork only if you want a customized, partially pruned Unsloth branch with specific fine-tuning fixes and you can absorb the maintenance risk. If you want the full upstream Studio feature set, this looks too divergent and behind to be the safer default.

Prefer this fork only if its specific fixes or simplifications match your environment and you can tolerate losing upstream breadth. For most users who want the full Studio/training/inference platform, upstream looks safer and more complete.

Prefer upstream unless you intentionally need this fork’s older 2024-era behavior. The fork looks stale and materially behind, but it may suit users who want a narrower, historical Unsloth variant centered on Gemma2, Ollama, GGUF, and LoRA fixes.

Prefer upstream unless you specifically need this fork’s custom behavior. This fork looks like a significantly diverged, stale branch that may offer useful local customizations, but it is a poor choice for adopters who want current features, broad model support, and active maintenance.

Prefer upstream unless you specifically need this older, modified baseline. This fork looks significantly behind and is best treated as a maintenance branch for legacy compatibility, not as the default choice for new adopters.

Prefer upstream unless you specifically need this fork's older training patches or want to maintain a reduced codebase yourself. For most adopters, the staleness and feature loss outweigh any simplicity benefit.

Choose the fork only if you need its older Llama-specific behavior. For most adopters, upstream is the better default because it is active and much richer in current training, inference, and Studio workflows.

Prefer this fork if you want a more tailored Studio experience around installation, GPU handling, and export/inference workflows. Prefer upstream if you want the newest fixes, broader parity, and lower upgrade risk.

Fork comparison

yangjianxin1/unsloth

38/100

stale

significant_divergence

Choose upstream unless you have a concrete reason to stay on an old, divergent fork. This fork looks like a stale 2024-era snapshot with substantial divergence, so it is better suited to niche compatibility or legacy workflows than to ongoing use.

Likely purpose

An older, specialized fork of Unsloth that looks centered on the core fine-tuning/runtime library rather than the much broader, actively maintained Studio and multi-platform workflow in upstream.

Best for

Users who specifically need an older Unsloth snapshot for compatibility or archival reasons.; Developers comfortable maintaining a divergent fork themselves.; People who only want the core fine-tuning library and do not need the current Studio/product surface.

Additional features

Early fork-era support and tuning for specific model families mentioned in the history, including Qwen2 and Phi-3.
A narrower training-library focus that may suit users who want the core Python fine-tuning stack without the newer Studio-heavy surface area.
Expands API, schema, or SDK surface area for downstream integrations and generated clients.
Introduces more operational workflow surface such as server-side handling, auth, session control, or admin tooling.
Changes agent workflow behavior, such as prompt handling, handoffs, worktrees, memory, or orchestration flow.

Missing features

It appears materially behind upstream, so newer Studio workflows such as the expanded web UI, multi-GPU device-map handling, GGUF export UI, and Windows installer fixes may be missing or incomplete.
Upstream’s newer inference and training conveniences around tool calling, code execution, data-recipe workflows, and broader platform polish are likely absent or harder to use here.
The fork is far behind upstream in commit history, so recent bug fixes and model-support updates from 2026 are not available.
It trails upstream by 201 commits, so some recent upstream features and fixes are likely not present yet.

Strengths

May be simpler for users who want the earlier core library behavior instead of a fast-moving full product stack.
Could be useful as a reference point for model-specific fine-tuning work around the 2024 timeframe.
Large divergence suggests it may contain custom local changes that matter to a niche workflow.

Risks

Very stale compared with upstream, so adopters should expect incompatibilities with current models, dependencies, and installer paths.
Large delete-heavy divergence makes maintenance and upstream syncing risky.
Missing recent upstream fixes increases the chance of Windows, packaging, and Studio regressions.
Because the fork is old, documentation and community support will likely lag behind current Unsloth.