RunsOnMy ModelsGPUs & MacsBy VRAMBest GPUCan I run…?OllamaStart hereAPI pricesVRAM calculatorNew

Can MiniMax M1 40K run on the H100 SXM?

Not fully in VRAM. Even at 2-bit with 8K context, MiniMax M1 40K needs about 163 GB, more than the 76.0 GB usable on the H100 SXM. It can still run with offload (slower): Q2_K with part of the model in system RAM, about 3.9–7.9 tokens/sec (estimate). Details and command.

Runs with offload (slower)

Runs with offload 2-bit (Q2_K) at 8K context: all attention layers plus the experts of 34 of 80 MoE layers on the H100 SXM (about 74.5 GB), and the experts of the other 46 layers in system RAM (about 87.3 GB). Needs a PC with at least 128 GB of system RAM. Estimated speed: 3.9–7.9 tokens/sec (slow but usable), versus a bandwidth ceiling of ~185 tok/s if it fit fully in VRAM.

llama-server -m minimax-m1-40k-Q2_K.gguf -c 8192 -ngl 999 --n-cpu-moe 46

Estimate, not a benchmark. No GGUF repo was found on Hugging Face yet; point -m at your own GGUF file. 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.

Memory by quant and context

Quant4K ctx8K ctx32K ctx128K ctxMax tok/s (est., 8K)
2-bit (Q2_K)161 GB No163 GB No170 GB No200 GB No—
3-bit (Q3_K_M)210 GB No211 GB No218 GB No248 GB No—
4-bit (Q4_K_M)262 GB No263 GB No270 GB No300 GB No—
5-bit (Q5_K_M)305 GB No306 GB No313 GB No343 GB No—
6-bit (Q6_K)350 GB No352 GB No359 GB No389 GB No—
8-bit (Q8_0)453 GB No455 GB No462 GB No492 GB No—
16-bit (F16/BF16)852 GB No853 GB No860 GB No890 GB No—

• real GGUF file size. Speed is a bandwidth ceiling (3350 GB/s), not a benchmark. Methodology.

Hardware that does fit it

Mac M3 Ultra 256GBMac M3 Ultra 512GB

What fits instead

"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 M1 40K 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-M1-40k), NVIDIA specs. Model data last checked 2026-11-18; refreshed hourly.