Downloads · 30 days
0
tritter-100m-bitnet
tzervas/tritter-100m-bitnet
tritter-100m-bitnet is a machine learning model from tzervas. Use it for the machine learning task on the model card, and read the license before you ship it in a product. The card lists the license as mit.
A 100M parameter BitNet b1.58 ternary-quantized model trained for code generation.
Access
Public
Updated Jan 29, 2026
Repo size
806 MB
Likes
0
Public
Repo makeup
Click a slice to open those files.
.pt403 MB · 100%
At a glance
- License
- mit
- Access
- Public
- Created
- Jan 29, 2026
- Updated
- Jan 29, 2026
- SHA
- 91adfb2b
Model card
From the Hugging Face model README
Tritter 100M BitNet
A 100M parameter BitNet b1.58 ternary-quantized model trained for code generation.
Training Details
- Parameters: 100,680,704
- Training tokens: 9,782,048
- Final loss: 11.2704
- Min loss: 11.0312
- Tokens/sec: 56136.5
- Training duration: 0:02:54.254825
- GPU: NVIDIA GeForce RTX 5080
Checkpoints
Intermediate checkpoints available at 10%, 20%, ..., 90% progress.
Generated with Tritter
- License
- mit
- Created
- Jan 29, 2026
- Updated
- Jan 29, 2026