# Open-Source-O1/Open-O1

Repository: https://github.com/Open-Source-O1/Open-O1
Canonical: https://ross.abutalabs.com/products/open-o1
Language: Python
License: Apache-2.0
License Family: permissive
Last push: 2024-11-21T02:47:37+00:00

## Health v2 (maintenance only)
Score: 22/100 (v2, computed 2026-09-02T17:46:02.011165+00:00)
- activity 0, release rhythm 35, longevity 50
- inputs: {"age_days": 700, "days_push": 650, "days_rel": null, "gap_med": null, "n_releases_24m": 0}
- flags: no_releases
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 1340, forks 54 (observed 2026-08-28T04:04:26.031693+00:00)

## What it is
Open O1 is an open-source effort to replicate the reasoning capabilities of OpenAI's proprietary O1 model by curating chain-of-thought SFT data and fine-tuning LLaMA and Qwen models with it. The project releases the resulting models and datasets on Hugging Face, targeting enhanced long-reasoning and problem-solving abilities in smaller open models.

## Use cases
- train an open-source model with o1-like reasoning
- download chain-of-thought SFT data for LLM fine-tuning
- fine-tune LLaMA or Qwen for long-reasoning tasks
- experiment with test-time scaling in open models
- find an open alternative to OpenAI o1
- study chain-of-thought activation data for reasoning

## When to choose
- you want open weights approximating o1-style reasoning
- you need CoT SFT data to fine-tune LLaMA/Qwen models
- you want to research test-time scaling and long reasoning in small models

## When to avoid
- you need production-grade o1-level performance
- you want a ready-made inference API rather than models to self-host
- you need a fully mature, actively maintained training framework

## Facets
- artifact type: dataset
- maturity: active
- function: llm-training, machine-learning, data-generation
- domain: large-language-models, deep-learning, machine-learning, artificial-intelligence
- platform: python, cloud
- tags: open-source-model, chain-of-thought, sft-data, llama, qwen, reasoning, test-time-scaling, huggingface, gpu

## Member repositories
- Open-Source-O1/Open-O1 (main) score 22

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:04:26.031693+00:00.
- Health v2: computed from the inputs above; adoption is never an input.
- Inferred fields (summary, facets, guidance): AI-extracted, prompt v1, taxonomy v1, on 2026-08-30T04:43:47.188543+00:00, confidence not recorded.
  - readme: https://github.com/Open-Source-O1/Open-O1 (fetched 2026-08-28T04:04:26.031693+00:00, sha 2e48da09acb5)
- Data as of 2026-08-30T08:39:29.467469+00:00.