MiniMax M3 — local AI model by MiniMax
Written by Jakub Rusinowski · Last updated
MiniMax M3 — a natively multimodal MoE at roughly 428B total and 23B active with a 1M-token context, under the minimax-community licence (June 2026). Multimodality is built into the single checkpoint: there is no separate vision release.
Variants
The smallest MiniMax M3 variant needs about 140 GB of VRAM at Q4_K_M — quantized weights plus framework overhead, before any KV cache.
| Model | VRAM at Q4 | VRAM | Context | Run it |
|---|---|---|---|---|
| MiniMax M3 428B-A23B → 428B (23B active) | ~259.2 GB | 1,000,000 | ollama run minimax-m3 | |
| MiniMax M3-VL 230B (10B active) | ~139.7 GB | 1,000,000 | ollama run minimax-m3 |
Memory is quantized weights plus overhead at Q4_K_M, from the same engine as the GPU & VRAM checker.
How to run MiniMax M3 locally
Install Ollama, then pull the tag.
ollama run minimax-m3Pick a size above for its own VRAM figure, speed estimate and install command.
Licence
Weights are downloadable and commercial use is permitted, subject to the licence’s acceptable-use terms.
Applies to: MiniMax M3 428B-A23BCommercial use permitted. No usage restrictions beyond attribution.
Applies to: MiniMax M3-VLRecommended GPU
The cheapest catalogued GPU that runs MiniMax M3 locally (min 140 GB VRAM) is the Apple M2 Ultra (192 GB).
Can I run MiniMax M3 on my GPU?
- MiniMax M3 on AMD Ryzen AI Max+ 395
- MiniMax M3 on Apple M1 Max
- MiniMax M3 on Apple M1 Ultra
- MiniMax M3 on Apple M2 Max
- MiniMax M3 on Apple M2 Ultra
- MiniMax M3 on Apple M3 Max
- MiniMax M3 on Apple M4 Max
- MiniMax M3 on Apple M5 Max
- MiniMax M3 on Apple M5 Pro
- MiniMax M3 on NVIDIA A100 80GB (PCIe)
- MiniMax M3 on NVIDIA DGX Spark
- MiniMax M3 on NVIDIA H100 80GB (PCIe)
- MiniMax M3 on NVIDIA RTX PRO 6000 Blackwell
MiniMax M3 — frequently asked questions
How much VRAM does MiniMax M3 need?
MiniMax M3 needs about 140 GB VRAM at Q4_K_M quantization for its smallest variant. Variants: MiniMax M3 428B-A23B (259 GB, Q4_K_M); MiniMax M3-VL (140 GB, Q4_K_M). On Apple Silicon, unified memory counts toward this requirement.
Can I run MiniMax M3 on an RTX 4090 (24 GB)?
MiniMax M3's smallest variant needs about 140 GB, which exceeds a single RTX 4090 (24 GB). Use multiple GPUs, a higher-VRAM card, or Apple Silicon with large unified memory.
What quantization should I use for MiniMax M3?
Q4_K_M is the best balance of quality and VRAM for MiniMax M3 in most cases. Choose Q8_0 for near-lossless quality if you have spare VRAM, or smaller quants (Q3/Q2) only when memory is tight.
How do I run MiniMax M3 with Ollama?
MiniMax M3 has no local Ollama tag — the published tag is cloud-hosted, so running it sends your prompts to a hosted GPU rather than your own machine.