Models / MiniMax M2.7
Best home rig for MiniMax M2.7
Legacy model2 measured resultsfastest plain decode: 29.5 tok/s
Newer model available: MiniMax M3.1 Flash. MiniMax M2.7 is a legacy model; its numbers stay here for reference. source ↗
Real, measured generation speeds for MiniMax M2.7 on home hardware, fastest first. Different quants/engines aren't directly comparable.
| Build | Quant | Engine | tok/s | Cost | Source |
|---|---|---|---|---|---|
| Minisforum MS-S1 MAX (Ryzen AI Max+ PRO 495, 192 GB): gpt-oss 120B at 55.8 tok/s | IQ4_XS | llama.cpp Vulkan | 29.5 | $7,399 (≈€6,602) | hommie.net ↗ |
| Minisforum MS-S1 MAX (Ryzen AI Max+ PRO 495, 192 GB): gpt-oss 120B at 55.8 tok/s | IQ4_XS, 32K context | llama.cpp Vulkan | 17.3 | $7,399 (≈€6,602) | hommie.net ↗ |
Rent before you buy: try this model on a cloud GPU for a few dollars first: Runpod (referral link; sign up with Google to get the new-user bonus)
FAQ
What is the fastest measured hardware for MiniMax M2.7?
Minisforum MS-S1 MAX (Ryzen AI Max+ PRO 495, 192 GB): gpt-oss 120B at 55.8 tok/s: 29.5 tok/s generation (llama.cpp Vulkan, IQ4_XS), from https://www.hommie.net/posts/ai-box.html.
Related: what GPU do you need for 2026 models · all builds