This is a minimal random-weight model derived from the Mistral NeMo architecture. It is published under the optimum-intel-internal-testing organization on Hugging Face and is intended for integration and performance testing of the Optimum Intel library. The model uses the Safetensors format and provides a chat template compatible with the Mistral NeMo tokenizer.
In the Foundation models & chat space, Tiny Random Mistral Nemo takes a focused approach. It focuses on testing and validating optimization tools for Intel hardware without using full-scale models. It is built as an open-source project for AI developers and researchers. The project is open source (Open Source). Tiny Random Mistral Nemo is available on the web and API.
Behind Tiny Random Mistral Nemo is Hugging Face, and it first shipped in 2025.
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