alignrl
LLM post-training playbook: SFT, GRPO, DPO, eval, and inference
- Latest release
- Mar 26, 2026
- Releases
- 3
- Known CVEs
- 0
- First release
- Mar 24, 2026
- License
- MIT
Insights
Activity
- Total releases
- 3
- Last 12 months
- 3
- Cadence
- ~daily
- Dependencies
- 19
Releases per month
last 12 monthsRelease mix
- minor 2
3
releases
Dependencies
Depends on
0.3.0-
accelerate
>=1.0
-
bitsandbytes
>=0.44
-
datasets
>=3.0
-
gradio
>=4.0
-
lm-eval
>=0.4
-
mlx-lm
>=0.18
-
mypy
>=1.10
-
peft
>=0.13
-
pydantic
>=2.0
-
pytest
>=8.0
1–10 of 19
Used by
Nothing tracked depends on this yet.
Releases
| Version | Released | |
|---|---|---|
0.3.0
minor
|
0.3.0
minor
Dependencies (19)
+ 11 more |
|
0.2.0
minor
|
0.2.0
minor
|
|
0.1.0
initial
|
0.1.0
initial
|