A community-quantized version of a large MiniMax language model (172B parameters with A10B active). It uses NVFP4 and GB10 quantization techniques to make the model more accessible for local or lower-resource inference. The model includes support for tool calling and is distributed on Hugging Face for use with the Transformers library and compatible inference engines.
In the Tool calling space, MiniMax M2.7 REAP 172B A10B NVFP4 GB10 takes a focused approach. It focuses on running very large language models with reduced memory footprint using specialized quantization. MiniMax M2.7 REAP 172B A10B NVFP4 GB10 is an open-source project aimed at developers. MiniMax M2.7 REAP 172B A10B NVFP4 GB10 is open source under the Apache-2.0 license. MiniMax M2.7 REAP 172B A10B NVFP4 GB10 is available on the web, the command line, and API.
Behind MiniMax M2.7 REAP 172B A10B NVFP4 GB10 is scottgl, and it first shipped in 2023. The project is developed in the open on GitHub with 11 stars and 2.4k commits in the last 90 days. Key capabilities include Large Language Model, Quantized Weights, and Tool Calling.
Summary written by a language model from the project’s public pages.
What PulseGate has recorded for this listing
Closest matches by what these projects do