Tenzro
Media generation

MiniMax H3 (Hailuo 3.0) H3-Base

MiniMax H3 omni-modal text-to-video with native stereo audio, 768p short edge, 4-15s. Licence excludes the EU, UK, South Korea and the USA, including outputs
MiniMax H3 Community Licensecommercial_customSource verified
Specification
Model ID
minimax-h3
Family
minimax-h3
Modality
Media generation
Parameters
H3-Base + Qwen3-VL-32B text encoder
Weights
134.1 GB
Minimum VRAM
144 GB
Capabilities
text2video
Pipeline
MiniMaxH3ModularPipeline
Backend
diffusers
Source

Weights and provenance.

Repository
MiniMaxAI/MiniMax-H3
Registry license
MiniMax H3 Community License
Source license
other
Access
Open — weights fetch without accepting additional terms.

The registry records this model as MiniMax H3 Community License while the source repository states other. Both are shown here rather than one being preferred. Treat the stricter of the two as binding until the discrepancy is resolved upstream.

Serve it
# Pull the weights onto a node
tenzro model download minimax-h3

# Serve it
tenzro model serve minimax-h3

# Or query the registry entry over JSON-RPC
curl https://rpc.tenzro.xyz \
  -X POST -H "Content-Type: application/json" \
  -d '{"jsonrpc":"2.0","id":1,
       "method":"tenzro_getModel",
       "params":[{"model_id":"minimax-h3"}]}'
Same family
MiniMax H3 FL2VA (GGUF Q4_K)
H3 FL2VA pruned Q4_K + Qwen3-VL-32B Q4_K_M
MiniMax H3 (Hailuo 3.0) H3-Base (int8)
H3-Base + Qwen3-VL-32B text encoder, int8
MiniMax H3 Ref2VA (FP8)
H3 Ref2VA FP8 + Qwen3-VL-32B INT8 text encoder
MiniMax H3 Ref2VA (GGUF Q4_K)
H3 Ref2VA pruned Q4_K + Qwen3-VL-32B Q4_K_M
← All models