Can MiniMax M3 run on the H100 SXM?
Not fully in VRAM. Even at 2-bit with 8K context, MiniMax M3 needs about 143 GB, more than the 76.0 GB usable on the H100 SXM. It can still run with offload (slower): Q3_K_M with part of the model in system RAM, about 6.1–12 tokens/sec (estimate). Details and command.
Runs with offload (slower)
Runs with offload 3-bit (Q3_K_M) at 8K context: all attention layers plus the experts of 22 of 58 MoE layers on the H100 SXM (about 72.7 GB), and the experts of the other 36 layers in system RAM (about 111 GB). Needs a PC with at least 128 GB of system RAM. Estimated speed: 6.1–12 tokens/sec (slow but usable), versus a bandwidth ceiling of ~367 tok/s if it fit fully in VRAM.
llama-server -hf bartowski/MiniMax-M3-GGUF:Q3_K_M -c 8192 -ngl 999 --n-cpu-moe 38 --no-mmproj
Estimate, not a benchmark. The command downloads the GGUF from Hugging Face. The speed range spans dual-channel DDR4-3200 to DDR5-6000 system RAM; your CPU, RAM speed and llama.cpp version move it. If llama.cpp runs out of GPU memory, raise --n-cpu-moe by 1-2; if VRAM is left over, lower it for more speed. Recent llama.cpp builds can also pick the split automatically. How we estimate offload.
+ 1.6 GB with image input. The figures on this page are for text only. With image input enabled, the vision projector (mmproj) file loads alongside the model. llama.cpp's -hf downloads and loads it automatically unless you add --no-mmproj. Calculator with image input on.
Memory by quant and context
| Quant | 4K ctx | 8K ctx | 32K ctx | 128K ctx | Max tok/s (est., 8K) |
|---|---|---|---|---|---|
| 2-bit (Q2_K) • | 143 GB No | 143 GB No | 146 GB No | 157 GB No | — |
| 3-bit (Q3_K_M) • | 184 GB No | 185 GB No | 188 GB No | 199 GB No | — |
| 4-bit (Q4_K_M) • | 245 GB No | 245 GB No | 248 GB No | 259 GB No | — |
| 5-bit (Q5_K_M) • | 286 GB No | 286 GB No | 289 GB No | 300 GB No | — |
| 6-bit (Q6_K) • | 345 GB No | 346 GB No | 349 GB No | 360 GB No | — |
| 8-bit (Q8_0) • | 424 GB No | 424 GB No | 427 GB No | 438 GB No | — |
| 16-bit (F16/BF16) | 797 GB No | 797 GB No | 800 GB No | 811 GB No | — |
• real GGUF file size from bartowski/MiniMax-M3-GGUF. Speed is a bandwidth ceiling (3350 GB/s), not a benchmark. Methodology.
Hardware that does fit it
Mac M2 Ultra 192GBMac M3 Ultra 256GBMac M3 Ultra 512GB
What fits instead
- No listed GPU with a list price runs MiniMax M3 fully in memory. The smallest listed hardware that does is the Mac M2 Ultra 192GB (192 GB).
- Largest popular model that fits the H100 SXM: gpt-oss 120B (117B, MoE) at F16, about 61.9 GB at 8K context.
"Runs" and "fits" mean the whole model in GPU memory at 8K context (4-bit or better for the model pick). Prices are US launch MSRPs, not today's street prices. "Popular" means the 60 most-downloaded models we track.
More
MiniMax M3 on every GPU and Mac · Every model that runs on the H100 SXM · Other models that fit 64 GB VRAM · Try other settings in the calculator
Guides: Best GPU for local LLMs by budget · Can I run gpt-oss, Llama 70B or Qwen3? · Mac vs PC for local LLMs
Data: Hugging Face Hub (MiniMaxAI/MiniMax-M3), NVIDIA specs. Model data last checked 2026-11-18; refreshed hourly.