What GPU do I need to run MiniMaxAI/MiniMax-H3?
A 33.1B language model for chat and instruction-following. 33.1B parameters, published in BF16. View on Hugging Face
MiniMax-H3 is published by MiniMaxAI on Hugging Face, with 5,532,597 downloads and 4,729 likes to date. It's a unlisted-architecture model built for image-text-to-video, published natively in BF16.
What MiniMax-H3 is
MiniMax-H3 is a 33.1B-parameter language model published by MiniMax on Hugging Face. It is released under Custom license.
License note: a lab-specific license (tagged "other" on Hugging Face); read the model card's own license section before commercial use. Facts in this section are sourced from MiniMax-H3's Hugging Face model card, not benchmarked by Aquanode.
What it's used for
- Chat assistants
- Instruction following
- Synthetic data generation
VRAM required & cheapest live GPU fit
Required VRAM = weight size at each precision times a flat 1.2 overhead for activations and fragmentation. The KV-cache grows with context and is not in that factor; it is listed per model below where the architecture is published. Full formula and assumptions: methodology.
| Precision | Weight size | Required VRAM | Cheapest live fit | GPUs needed | Est. $/hr (full fit) |
|---|---|---|---|---|---|
| BF16 | 61.7 GB | 74.0 GB | RTX PRO 6000 | 1 | $1.21/hr |
| cheaper alt. | RTX A5000 | 4 | $0.704/hr | ||
| FP8 (quantized) | 30.8 GB | 37.0 GB | RTX 4090 | 1 | $0.662/hr |
| cheaper alt. | RTX 4000 SFF Ada | 2 | $0.396/hr | ||
| INT4 (quantized) | 15.4 GB | 18.5 GB | RTX A5000 | 1 | $0.176/hr |
A GPU is only matched to a row if its hardware supports that precision, and the primary recommendation is always a single-GPU fit when one exists.
INT4 requires a quantized checkpoint actually published for this model, check its Hugging Face page before relying on this row.
Cheapest way to run MiniMax-H3 at its published (BF16) precision: 1× RTX PRO 6000, at $1.21/hr per GPU ($1.21/hr total). Quantizing to FP8 or INT4 (rows above) can cost less, but requires a compatible quantized checkpoint to exist for this model.
MiniMax-H3: common questions
Can MiniMax-H3 run on a single GPU?
Yes, but not on a desktop card. At BF16 it needs 74.0 GB of VRAM, more than a single 32 GB desktop card holds. The cheapest live single-card fit is a 95.0 GB RTX PRO 6000 at $1.21/hr.
What is the least VRAM MiniMax-H3 can run in?
18.5 GB, at INT4 (quantized), which fits a 24 GB card, against 74.0 GB at BF16. That row assumes an INT4 checkpoint has actually been published for this model, so check its Hugging Face page before planning around it.
Does quantizing MiniMax-H3 lower the GPU bill?
Yes. At BF16 the cheapest live fit is one RTX PRO 6000 at $1.21/hr. At INT4 (quantized) it drops to one RTX A5000 at $0.176/hr, provided a quantized checkpoint exists for it.
How to run MiniMax-H3
Run MiniMax-H3 with SGLang
From MiniMaxAI/MiniMax-H3's own deployment docs.
# Original checkpoint, both task families (SGLang, vLLM):
hf download MiniMaxAI/MiniMax-H3 --include "model_index.json" "FL2VA/*" "Ref2VA/*" --local-dir MiniMax-H3
# Or a single task family:
hf download MiniMaxAI/MiniMax-H3 --include "model_index.json" "FL2VA/*" --local-dir MiniMax-H3Source: https://huggingface.co/MiniMaxAI/MiniMax-H3/raw/main/README.md
Run MiniMax-H3 with GGUF quantizations
Prebuilt GGUF weights published at unsloth/MiniMax-H3-GGUF. Run with llama.cpp's llama-server or load the repo directly in LM Studio.
llama-server -hf unsloth/MiniMax-H3-GGUFDeploy MiniMax-H3 on Aquanode
Aquanode has no one-click deploy template for MiniMax-H3; it comes with ComfyUI preinstalled, so you only need to load the checkpoint, not install anything. Aquanode sells GPU pods billed per second, not a hosted inference API.
- Launch the ComfyUI template sized to the requirement above (1× RTX PRO 6000 or larger).
- Open a terminal on the pod, or save one of the commands above as a startup script so it runs automatically the first time the pod boots.
- Run the command and connect to the resulting endpoint.
Launch the ComfyUI templateWeight-to-VRAM math, the fit rules, and how live prices are normalized: full methodology.
More MiniMax H3 models
- FastVideo-Minimax-FastH3-Preview-v0.2 (35.0B, BF16)
- FastVideo-FastH3-4-step-Preview-v1-VSA-DataFree (35.0B, BF16)
Lineage
Quantized and fine-tuned versions
More on MiniMax-H3
Fits on an 80 GB GPU at BF16: every model that fits in 80 GB.
Fits on a 48 GB GPU at FP8: every model that fits in 48 GB.
Fits on a 24 GB GPU at INT4: every model that fits in 24 GB.
Related reading: RTX PRO 6000 pricing and specs, How much VRAM you need for LLMs, and What AI inference is.
