Documentation Navigation

Models & pricing / Models / MiniMax-H3

MiniMax-H3 Overview

This model's data has not been fully verified and benchmarks may be missing.

MiniMax-H3 is a multimodal-general AI model created by MiniMaxAI, featuring 33.1B parameters and a context window of unknown.

MiniMax-H3 is a 33.1B-parameter image text to video model developed by MiniMaxAI. Released on 2026-07-28 with 2,677 likes and 12,102 downloads on Hugging Face.

Model Lineage & Specification

minimax-h3 is MiniMaxAI's primary release in the current family.

DeveloperMiniMaxAI
API Identifierminimax-h3
Parameters33.1B
Context Windowunknown
LicenseUnknown

Comparable models

MiniMax-H3

MiniMax-H3 is a 33.1B-parameter image text to video model developed by MiniMaxAI...

API IDminimax-h3
Typeopen weights
Parameters33.1B
Contextunknown

Muse-Glimmer-30B-GGUF

Muse-Glimmer-30B-GGUF is a undisclosed-parameter image text to text model develo...

API IDmuse-glimmer-30b-gguf
Typeopen weights
Parameters29.6B
Context131,072+

Minimax-H3-nvfp4-INT4-INT8-Convrot

Minimax-H3-nvfp4-INT4-INT8-Convrot is a undisclosed-parameter image text to vide...

API IDminimax-h3-nvfp4-int4-int8-convrot
Typeopen weights
Parameters
Contextunknown

BigBang-v1

SGLang is a high-performance serving framework for large language models and mul...

API IDbigbang-v1
Typeopen weights
Parameters36.0B
Contextunknown

MiniMax-H3 is a 33.1B-parameter image text to video model developed by MiniMaxAI. Released on 2026-07-28 with 2,677 likes and 12,102 downloads on Hugging Face.