Skip to content
localmodel.run

Text model · MiniMax

MI MiniMax-M1-80k: RAM and VRAM requirements

MiniMax-M1-80k needs about 284.6 GB to run at Q4_K_M (the Q4_K_M GGUF file is ~278.7 GB to download; KV cache and overhead add the rest), or about 490.4 GB at Q8_0. The lightest hardware that runs it is a high-memory machine.

MiniMax family · 456B params (Mixture-of-Experts: activates only 45.9B of 456B params per token, so generation is faster than the total size suggests) · released Jun 2025.

Shopping for hardware? See what runs MiniMax-M1-80k →

License Apache-2.0 · Commercial OK ↓ 628/mo ♥ 692 on HuggingFace
Q4_K_M GGUF
278.7 GB
Q8_0 GGUF
-
Memory @ Q4 (4k)
~284.6 GB
Context
1000 k

Q4_K_M and Q8_0 are GGUF download (file) sizes. Memory @ Q4 includes the KV cache and overhead, so it runs larger than the file on disk.

Will it run on your device?

MiniMax-M1-80k runs on 0 of 40 tracked devices at Q4_K_M.

0 run well 0 tight fit 40 too small
Fit check Q4_K_M
No, not enough memory
needs 284.6 GB usable 10.5 GB
See the full breakdown

Memory breakdown

Weights (Q4_K_M)278.7 GB
+
KV cache (4k)5.1 GB
+
Overhead0.8 GB
=
Total284.6 GB

How context length changes it

4k context ~284.6 GB 32k context ~320.5 GB 128k context ~443.5 GB

Longer context grows the KV cache, which for MiniMax-M1-80k is sized by its 45.9B active params, not the full 456B. It needs ~284.6 GB at 4k and ~443.5 GB at 128k.

Quantization sizes

GGUF quantson disk
Quantization Size on disk
Q2_K 191 GB est
Q3_K_M 222.9 GB est
Q4_K_M (default) 278.7 GB est
Q5_K_M 324.9 GB est
Q6_K 373.9 GB est
Q8_0 484.5 GB est
FP16 912 GB est

Lower quant = smaller and faster, slightly lower quality. Q4_K_M is the common default.

Run it

Run with a GGUF-compatible app such as PocketPal AI on mobile.

Which devices can run MiniMax-M1-80k?

Apple Silicon Macs

RAM-only laptops

iPhone & iPad

Android

NVIDIA GPUs

AMD GPUs

Same job, different size

Pick by what fits your memory: step down to free up VRAM, or step up if you have headroom.

Next size down MiniMax M3 428B Next size up Qwen3-Coder 480B-A35B Instruct 480B

Similar models

Head-to-head

FAQ

How much VRAM or RAM does MiniMax-M1-80k need?

At Q4_K_M, MiniMax-M1-80k needs about 284.6 GB (weights ~278.7 GB + KV cache + overhead) at a 4k context. At Q8_0 budget ~490.4 GB.

What is the Q4_K_M GGUF file size of MiniMax-M1-80k?

The Q4_K_M GGUF file is about 278.7 GB to download, and the Q8_0 GGUF is about 484.5 GB. That is the weights file on disk; to run it you also need room for the KV cache and overhead, so budget ~284.6 GB of memory at Q4_K_M.

Can MiniMax-M1-80k run on a laptop?

MiniMax-M1-80k is large; you need a high-memory Mac or multi-GPU setup at Q4_K_M.

Is MiniMax-M1-80k cheaper to run because it is a MoE model?

It is faster, not lighter. MiniMax-M1-80k activates only 45.9B of 456B params per token (so it runs quickly), but all experts must stay in memory, so it still needs memory for the full 456B.

Can I use MiniMax-M1-80k commercially?

Yes. MiniMax-M1-80k is licensed Apache-2.0, which permits commercial use.

Understand the numbers

Short guides to the ideas behind MiniMax-M1-80k's memory and quant figures.

MiniMax's 456B reasoning MoE (45.9B active) with a 1M context. Open weights, but no prebuilt Q4 GGUF yet, so the size shown is computed from the formula.

Sources

Last validated 2026-08-03. Memory figures are estimates. See methodology.