Run MiniMax M3 locally
MiniMax M3 is 427B parameters (23B active per token — a mixture-of-experts, so it still needs room for all 427B in memory). Here's the VRAM it needs at each quantization, the smallest GPU that fits, and how a self-hosted box compares with the cheapest verified API.
VRAM by quantization
| Quantization | VRAM needed | Fits on | Notes |
|---|---|---|---|
| int4 (Q4) | 256.2 GB | 4× H100 (320 GB) ($6.92/hr) | smallest footprint, minor quality loss |
| int8 (Q8) | 512.4 GB | 8× H100 (640 GB) ($13.84/hr) | near-lossless |
| fp16 (full) | 1024.8 GB | multi-node (>640 GB) | reference quality |
VRAM ≈ params × bytes/param × 1.2 (overhead). See the fullmethod and break-even calculator.
Self-host vs API
Cheapest API for MiniMax M3 is $1.19 / 1M tokens(blended) via CoreWeave. Running it yourself in int4 fits a4× H100 (320 GB) at $6.92/hr — about $5,051.6/month at 24/7. Those two lines cross at roughly 4.2B tokens/month: below that the API wins on cost, above it the dedicated GPU does (assuming you keep it busy). Tune your own volume in the break-even calculator.
Run it at home
In int4, MiniMax M3 needs 256.2 GB — more than any consumer desktop or Apple machine we track holds, so at home it means a multi-GPU server. Renting cloud GPUs by the hour, or the API, is the practical route. See the machines table for the ceiling.
Get the weights
Quantized builds on the model hubs — links search live, so they track new builds as they appear:
- GGUF builds llama.cpp · Ollama · LM Studio
- AWQ builds GPU serving (vLLM)
- GPTQ builds GPU serving
- MLX builds Apple silicon
- Ollama library one-command local run
License: open weights under MiniMax Community License · terms. Compare all providers on the minimax-m3 API page.