What are the hardware requirements for K2 Horizon MoVA 36B A4B FP8?

K2 Horizon MoVA 36B A4B FP8 requires approximately 40.6 GB of VRAM and supports a context window of 512K tokens. Quantized variants may run on less VRAM; see the Quantized Models section on this page.

Who developed K2 Horizon MoVA 36B A4B FP8 and how large is it?

K2 Horizon MoVA 36B A4B FP8 is developed by IFM, a model with 36b parameters. The model is published as open weights on Hugging Face and indexed on LLM Explorer with full benchmark history.

Where can I download or evaluate K2 Horizon MoVA 36B A4B FP8?

K2 Horizon MoVA 36B A4B FP8 is hosted on Hugging Face and linked from this page. LLM Explorer also lists quantized variants and similar alternatives if available.

K2 Horizon MoVA 36B A4B FP8 by IFM — VRAM 40.6GB, 512K context

Name: K2 Horizon MoVA 36B A4B FP8
Author: IFM

K2 Horizon MoVA 36B A4B FP8 is an open-source language model by IFM. Features: 36b LLM, VRAM: 40.6GB, Context: 512K, License: apache-2.0, LLM Explorer Score: 0.34.

36b Base model:ifm/k2-horizon-mova... Base model:quantized:ifm/k2-ho... Conversational Custom code Dataset:ifm/k2-horizon-midtrai... Dataset:ifm/k2-horizon-pretrai... En Fp8 Ifm K2-horizon K2 horizon Moe Mova Open-weights Region:us Safetensors Sharded Tensorflow

Model Card on HF 🤗: https://huggingface.co/IFM/K2-Horizon-MoVA-36B-A4B-FP8

K2 Horizon MoVA 36B A4B FP8 Parameters and Internals

LLM Name	K2 Horizon MoVA 36B A4B FP8
Repository 🤗	https://huggingface.co/IFM/K2-Horizon-MoVA-36B-A4B-FP8
Model Name	K2-Horizon-MoVA-36B-A4B-FP8
Base Model(s)	K2 Horizon MoVA 36B A4B IFM/K2-Horizon-MoVA-36B-A4B
Model Size	36b
Required VRAM	40.6 GB
Updated	2026-09-05
Maintainer	IFM
Model Type	k2_horizon
Model Files	0.2 GB: 1-of-48 0.2 GB: 2-of-48 0.2 GB: 3-of-48 1.0 GB: 4-of-48 1.0 GB: 5-of-48 1.0 GB: 6-of-48 1.0 GB: 7-of-48 1.0 GB: 8-of-48 1.0 GB: 9-of-48 1.0 GB: 10-of-48 1.0 GB: 11-of-48 1.0 GB: 12-of-48 1.0 GB: 13-of-48 1.0 GB: 14-of-48 1.0 GB: 15-of-48 1.0 GB: 16-of-48 1.0 GB: 17-of-48 1.0 GB: 18-of-48 1.0 GB: 19-of-48 1.0 GB: 20-of-48 1.0 GB: 21-of-48 1.0 GB: 22-of-48 1.0 GB: 23-of-48 1.0 GB: 24-of-48 1.0 GB: 25-of-48 1.0 GB: 26-of-48 1.0 GB: 27-of-48 1.0 GB: 28-of-48 1.0 GB: 29-of-48 1.0 GB: 30-of-48 1.0 GB: 31-of-48 1.0 GB: 32-of-48 1.0 GB: 33-of-48 1.0 GB: 34-of-48 1.0 GB: 35-of-48 1.0 GB: 36-of-48 1.0 GB: 37-of-48 1.0 GB: 38-of-48 1.0 GB: 39-of-48 1.0 GB: 40-of-48 1.0 GB: 41-of-48 1.0 GB: 42-of-48 1.0 GB: 43-of-48
Supported Languages	en
Model Architecture	K2HorizonForCausalLM
License	apache-2.0
Context Length	524288
Model Max Length	524288
Transformers Version	5.13.0
Vocabulary Size	250624
Torch Data Type	bfloat16

Rank the K2 Horizon MoVA 36B A4B FP8 capabilities

Have you tried this model? Rate its performance — this feedback helps the ML community find the right model for their needs.

Instruction Following and Task Automation
Factuality and Completeness of Knowledge
Censorship and Alignment
Data Analysis and Insight Generation
Text Generation
Text Summarization and Feature Extraction
Code Generation
Multi-Language Support and Translation

Best Alternatives to K2 Horizon MoVA 36B A4B FP8

Best Alternatives	Context / RAM	Downloads	Likes
K2 Horizon MoVA 36B A4B	512K / 63 GB	433	139

Note: green Score (e.g. "73.2") means that the model is better than IFM/K2-Horizon-MoVA-36B-A4B-FP8.