NVIDIA/Model-Optimizer

Model-Optimizer

✅ Recently maintained

A unified library of SOTA model optimization techniques like quantization, distillation, pruning, neural architecture search, speculative decoding, etc. It compresses deep learning models for downstream deployment frameworks like TensorRT-LLM, TensorRT, vLLM, etc. to optimize inference speed.

5,493Stars
788Forks
+517-day stars

Python · Apache-2.0 · Last push 2026-11-02

Official repository & installation ↗

Star history

2025-09-292026-11-02
View historical data
DateStars
2026-10-045,192
2026-10-055,213
2026-10-065,219
2026-10-075,229
2026-10-085,237
2026-10-095,244
2026-10-105,253
2026-10-115,265
2026-10-125,284
2026-10-135,294
2026-10-145,311
2026-10-155,328
2026-10-165,338
2026-10-175,347
2026-10-185,359
2026-10-195,371
2026-10-205,384
2026-10-215,400
2026-10-225,417
2026-10-235,427
2026-10-245,430
2026-10-255,440
2026-10-265,447
2026-10-275,460
2026-10-285,465
2026-10-295,473
2026-10-305,474
2026-10-315,479
2026-11-015,490
2026-11-025,493

Get the project

git clone https://github.com/NVIDIA/Model-Optimizer.git

Related projects & alternative candidates

Related domains do not imply feature parity. Review official documentation and licenses.