BI

bitsandbytes

Model Optimization·infrastructure·open·#123 of 884·-3·Stable

82.7

Strong

High confidence
WebsiteGitHub

Lightweight CUDA library for 8-bit and 4-bit quantization of large language models.

Pillar Breakdown

Adoption

35%

86.8

Maintenance

30%

81.8

Friction

20%

98.1

Ecosystem

15%

66.7

Momentum

0.42Stable
7d change -0.46
High confidence

In Model Optimization

Ranked #4 of 17

1onnx
87.8
2Safetensors
83.6
3ONNX Runtime
82.7
4bitsandbytes
82.7
5DeepSpeed
78.4
6TensorRT
75.9
7Optimum
75.6
8numba
74.8
See all 17 in Model Optimization →

Similar Tools

OR

ONNX Runtime

Model Optimization

82.7

SA

Safetensors

Model Optimization

83.6

DE

DeepSpeed

Model Optimization

78.4

ON

onnx

Model Optimization

87.8