fermion-research
Five-value sub-2-bit LLMs: chat with a ~2 GB 8B container at native-runtime speed, load it as a Transformers model, or serve it on an OpenAI-compatible endpoint
Activity
- Latest release
- 1d ago
- Total releases
- 9
- Cadence
- ~daily
- Last 12 months
- 9
Details
- License
- Apache-2.0
- First release
- Jul 27, 2026
| Version | Released | |
|---|---|---|
0.1.13
patch
|
0.1.13
patch
Dependencies (4)
|
|
0.1.12
patch
|
0.1.12
patch
Dependencies (4)
|
|
0.1.11
patch
|
0.1.11
patch
Dependencies (4)
|
|
0.1.10
patch
|
0.1.10
patch
Dependencies (4)
|
|
0.1.9
patch
|
0.1.9
patch
Dependencies (4)
|
|
0.1.8
patch
|
0.1.8
patch
Dependencies (4)
|
|
0.1.7
patch
|
0.1.7
patch
Dependencies (4)
|
|
0.1.6
patch
|
0.1.6
patch
Dependencies (4)
|
|
0.1.5
initial
|
0.1.5
initial
Dependencies (4)
|