MiniMax-M3.1-Flash
MiniMax-M3.1-Flash is an open-weights language model from MiniMax, released October 19, 2026. It has 220B parameters (mixture-of-experts, 2.2B active). The context window is 1M tokens. API price: $0.15 input / $0.6 output per million tokens.
Check it against your own PC →
Can you run MiniMax-M3.1-Flash locally?
The smallest common GPU it runs on is 32 GB of VRAM (with 32 GB of RAM): IQ1_M, GPU + experts in RAM, roughly 190 tokens per second. Estimates assume 32 GB of DDR5 RAM, a 16K context and llama.cpp-style GGUF quantization. Enter your exact hardware in the live app.
| GPU memory | Verdict | Best quant | VRAM | RAM | tok/s |
|---|---|---|---|---|---|
| 6 GB VRAM | Too large | – | – | – | – |
| 8 GB VRAM | Too large | – | – | – | – |
| 12 GB VRAM | Too large | – | – | – | – |
| 16 GB VRAM | Too large | – | – | – | – |
| 24 GB VRAM | Too large | – | – | – | – |
| 32 GB VRAM | Runs · GPU + CPU experts | IQ1_M (experimental) | 31.2 GB | 19.6 GB | ~190 |
Ready-made quantizations on Hugging Face: peasantsmith/MiniMax-M3.1-Flash-Maya-GGUF · unsloth/MiniMax-M3.1-Flash-GGUF · LyJonathan/MiniMax-M3.1-Flash-GGUF · Baekpica/MiniMax-M3.1-Flash-VQ-GGUF · AtomicChat/MiniMax-M3.1-Flash-GGUF · orcarouter/MiniMax-M3.1-Flash-abliterated-GGUF
Benchmark results
Epoch AI has not published results for MiniMax-M3.1-Flash yet. New models are usually evaluated within days to weeks.
Links
Hugging Face: MiniMaxAI/MiniMax-M3.1-Flash
More from MiniMax
MiniMax-M3 · MiniMax-M2.7 · MiniMax-M2.5 · MiniMax-M2.1 · MiniMax-M2 · MiniMax-M1-80k · MiniMax-M1-40k