microsoft/BitNet
BitNet
Official inference framework for 1-bit LLMs
Builder
Microsoft
microsoft • big-tech
Stars
37,052
Using upstream star count
Forks
3,254
Using upstream fork count
Open Issues
0
Activity Score
0/100
0 commits in 30d
Created
Aug 5, 2024
Project creation date
README Summary
BitNet is Microsoft's official inference framework specifically designed for 1-bit Large Language Models (LLMs). The framework provides optimized tools and utilities to run inference on extremely quantized neural networks where weights are reduced to 1-bit representations, enabling significant memory and computational savings.
AI Dev Skills
Unmapped
Neural Network QuantizationLarge Language Model Inference1-bit Weight QuantizationModel CompressionEfficient Deep LearningLow-precision Neural NetworksTransformer ArchitectureMemory-efficient Computing
Tags
Neural Network QuantizationLarge Language Model Inference1-bit Weight QuantizationModel CompressionEfficient Deep LearningLow-precision Neural NetworksTransformer ArchitectureMemory-efficient ComputingEfficient AIEdge/MobileTextOn-device AIEdge AI Language ModelsSelf-hostedWeight QuantizationOn-premiseMemory-efficient Text GenerationLow-power Natural Language ProcessingNeural Network OptimizationSmall Language ModelsLow-precision ComputingResource-constrained LLM Deployment1-bit Neural Network QuantizationPython
Taxonomy
Deployment Context
Modalities
Skill Areas
Recent Activity
Updated 1 months ago
7 Days
0
30 Days
0
90 Days
0
Quality
research- Quality
- medium
- Maturity
- research
Categories
Inference & ServingPrimaryNLP & TextEdge & Mobile AIOther AI / MLFoundation Models
PM Skills
Scale & Reliability
Languages
Python100.0%
Timeline
- Project created
- Aug 5, 2024
- Forked
- Mar 12, 2026
- Your last push
- 1 months ago
- Upstream last push
- 1 months ago
- Tracked since
- Mar 10, 2026
Similar Repos
pgvector cosine similarity · $0
Loading…