Kimi-K2.5-NVFP4 is an open-weight multimodal foundation model published by NVIDIA on Hugging Face. It supports text, image, video, and tool-calling capabilities through a custom chat template. The model is designed for local inference and developer integration via the Hugging Face ecosystem and Transformers library.
In the Multimodal & vision space, Kimi K takes a focused approach. It focuses on accessing and running a capable open-weight multimodal language model locally or on custom infrastructure. Kimi K is an open-source project aimed at developers. Kimi K is open source under the Apache-2.0 license. It ships for the web and API.
NVIDIA builds and maintains Kimi K, and it first shipped in 2024. Development happens publicly on GitHub with 3.3k stars and 354 commits in the last 90 days. Among its 4 catalogued features are Multimodal Input, Tool Calling, and Chat Template. It exposes integrations via a public API.
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