K2-Horizon-MoVA-36B-A4B

Explore

Apps & Tools

K2-Horizon-MoVA-36B-A4B

High-performance MoE vision-language model from the Institute of Foundation Models.

byInstitute of Foundation Models

PlayableOpen source

K2-Horizon-MoVA-36B-A4B media is blocked

Allow external media to connect to the provider and play this content.

Description

K2-Horizon-MoVA-36B-A4B is a 36B parameter Mixture-of-Experts vision-language model. It features a MoVA architecture and has been benchmarked with a score of 25 on the Artificial Analysis Intelligence Index, supported by vLLM and SGLang.

Descriptions, tags, and model credits may be AI-generated or inferred from public sources and can be incomplete or wrong. Learn more.

Details

Why this is interesting: A significant MoE vision-language model release from IFM, achieving a score of 25 on the Artificial Analysis Index.
Attributed to: Institute of Foundation Models@IFM_AI
Built with: K2 Horizon MoVA 36B-A4B Qwen3.6 Redwood AI
Links: huggingface.co
View tweet
View GitHub repo
Tags: #multimodal #moe #llm #vision-language

Discovered: X
First posted: X · Sep 22, 2026
Listed: Added 1h ago
Build: Strong
Models: Strong