MiniCPM5-1B NVFP4
NVFP4 weight-only quantization of openbmb/MiniCPM5-1B using llm-compressor / compressed-tensors.
Best runtime target: Blackwell-class NVIDIA GPUs with vLLM support for NVFP4.
Licensing and attribution
This model distribution is licensed under the Apache License, Version 2.0. See LICENSE. Existing third-party copyright, license, and attribution notices remain applicable.
- Upstream: openbmb/MiniCPM5-1B; declared license:
apache-2.0.