Models / MiniMax M3.1 Flash
Best home rig for MiniMax M3.1 Flash
Current model428B total · 23B active1 measured resultfastest plain decode: 25.3 tok/s model info ↗
Will it fit my GPU? Best value GPU per euro →
Real, measured generation speeds for MiniMax M3.1 Flash on home hardware, fastest first. Different quants/engines aren't directly comparable.
| Build | Quant | Engine | tok/s | Cost | Source |
|---|---|---|---|---|---|
| Mac Studio M3 Ultra (60-core GPU) 256 GB running MiniMax M3.1 Flash | oQ3e-mtp (MTP off) | oMLX v0.7.2 | 25.3 | omlx.ai ↗ |
Memory math: how much VRAM / RAM does MiniMax M3.1 Flash need?
MiniMax M3.1 Flash has 428B parameters, about 23B active per token (source: Hugging Face model card: ~428B total, 23B activated per token (quantization-aware trained)). Weight size = parameters × bits-per-weight ÷ 8, before context (KV cache) and runtime overhead.
| Format | Bits/weight | Weights | Smallest common memory size that holds the weights alone |
|---|---|---|---|
| Q8_0 | 8.50 | 454.8 GB | 512 GB |
| Q6_K | 6.56 | 351.1 GB | 512 GB |
| Q5_K_M | 5.70 | 305.1 GB | 512 GB |
| Q4_K_M | 4.89 | 261.9 GB | 512 GB |
| Q3_K_M | 4.00 | 213.8 GB | 256 GB |
| Q2_K | 3.16 | 169.0 GB | 192 GB |
Bits-per-weight for GGUF quants are llama.cpp's own figures (llama.cpp quantize README). "As measured" rows apply the source's stated bits-per-weight to the parameter count; real files differ slightly because some layers stay at higher precision. Try other combinations in the VRAM calculator.
Why it can still run on smaller GPUs: it is a mixture-of-experts model, so only about 23B of its 428B parameters are used for each generated token. Runtimes such as llama.cpp (--n-cpu-moe), ExLlamaV3 or Project Maya can keep most expert weights in system RAM and only the shared layers and hot experts in VRAM. Then you need VRAM + system RAM ≥ weights, and speed is limited mainly by RAM bandwidth.
Measured on setups with less GPU memory than the Q4_K_M weights (so part of the model lived in system RAM or was streamed from disk):
- Apple M3 Ultra (60-core GPU, unified) (256 GB) + 256 GB unified: 25.3 tok/s (oQ3e-mtp (MTP off))
FAQ
How much VRAM does MiniMax M3.1 Flash need?
The weights are about 261.9 GB at Q4_K_M (428B params × 4.89 bits ÷ 8, before context/KV cache). Add memory for context (KV cache) and the runtime. The smallest-memory setup with a measured result here is Apple M3 Ultra (256 GB) at 25.3 tok/s (oQ3e-mtp (MTP off)). At Q4_K_M the weights are about 261.9 GB, more than 256 GB, so Q4_K_M doesn't fully fit without offloading part of the model to system RAM.
What is the fastest measured hardware for MiniMax M3.1 Flash?
Apple M3 Ultra: 25.3 tok/s generation (oMLX v0.7.2, oQ3e-mtp (MTP off)), from https://omlx.ai/benchmarks/performance/m5fqe90r.
Which GPUs have measured results for MiniMax M3.1 Flash?
Apple M3 Ultra.
Can it run on… Apple M3 Ultra
Related: what GPU do you need for 2026 models · all builds