bytedance / axion1_0
Frontier-level agent capabilities with nothing more than high-quality data and a simple training recipe: 2 SFT + RL, no mid-training/DPO/OPD. Full recipe released.
Why this job: topics: multi-agent, agentic-workflow, ai-agents, agents, agentic-ai
What it does
- Three-stage training. Two stages of supervised fine-tuning (SFT) followed
- Scientific reasoning and tool use. Axion 1.0-122B scores 59.0 on full
- Verifiable training data. Reasoning problems are strengthened using
- Open training code. The repository includes data generation and
Get started
No single install command. Read the steps in the README (opens in new tab)
- Maintenance
- Active
- Last commit 3 days ago
- Stars
- 63
- +49 this week
- Licence
- Apache-2.0
- Allows business use
- Downloads for
- None listed
- No release files
- Latest version
- None yet
- No releases
- Built with
- Python
- 98% of the code
Is it looked after?
- Contributors
- 2
- Last 90 days
- Issues
- 0% closed
- 2 still open
- Pull requests
- 1 merged
- 0 waiting
- Releases
- None
- Never published
- No contributing guide
- No security policy
- No code of conduct
Similar repos
| Repo | Stars | Last commit | Licence | Downloads for |
|---|---|---|---|---|
| neuron-core / neuron-laravel | 123 | 7 days ago | MIT | No downloads |
More: languages, topics and project record
Languages
- Python 98%
- Shell 2%
Topics
- llm
- agentic-ai
- agentic-workflow
- ai-agents
- multi-agent
- agent-skills
- agents
- ai
Project record
- JobAgent frameworks
- StatusNew to our list
- Commits37
- Forks6
- Watching0
- CreatedNov 2, 2026
- First seen by usNov 7, 2026
- CheckedNov 9, 2026