NC

neural-compressor

Model Optimization·infrastructure·open·#488 of 944·-5·Rising·Breakout #14

70.7

Moderate

High confidence
GitHub

SOTA low-bit LLM quantization (INT8/FP8/MXFP8/INT4/MXFP4/NVFP4) & sparsity; leading model compression techniques on PyTorch, TensorFlow, and ONNX Runtime

Pillar Breakdown

Adoption

35%

60.2

Maintenance

30%

81.7

Friction

20%

98.8

Ecosystem

15%

54.4

Momentum

0.50Rising
7d change 0.00
High confidence

In Model Optimization

Ranked #8 of 16

1onnx
87.2
2Safetensors
84.1
3bitsandbytes
82.8
4ONNX Runtime
81.4
5DeepSpeed
78.0
6Optimum
77.4
7TensorRT
76.8
8neural-compressor
70.7
See all 16 in Model Optimization →

Similar Tools

LL

lit-llama

Model Optimization

67.3

TO

torchdistill

Model Optimization

66.4

LT

lightly-train

Model Optimization

66.3

TE

TensorRT

Model Optimization

76.8