MiniMax M1 40k — Hardware Requirements & GPU Compatibility
ChatFunctions
MiniMax M1 40k is a 456.1B-parameter open language model from MiniMax in the MiniMax family. It supports a context window of up to 10,240,000 tokens. At BF16 it needs about 912.98 GB of VRAM — see which GPUs and Macs can run it below.
MiniMax M1 40k runs on a multi-GPU server
BF16 needs ~916.5 GB · about 8× H200 SXM 141 GB
Apache 2.0
Not set up for it?
Reference GPU · RTX 4090 24 GB
Estimates, ±20% · how we estimate
Run MiniMax M1 40k
One recommended way. Every other method, OS and quant is in the step-by-step guide.Each way opens a step-by-step guide for your OS and quant.
Recommended
vllm serve MiniMaxAI/MiniMax-M1-40k --max-model-len 16384 --tensor-parallel-size 8Size the nodes from the table above
Download: Weights on Hugging Face
Ways to runOther ways
each opens step-by-stepRun in cloud
Rent 8× H200 SXM node · BF16
- Estimated hourly price
- ~$28.72/hr
- est., whole server
- GPUs
- 8 GPUs
- 1.1 TB of VRAM
$5+ bonus for new users
How Much VRAM Does MiniMax M1 40k Need?
Select a quantization level; the listed hardware adheres to it.
| Quantization | VRAM | +32K | Runs on | |
|---|---|---|---|---|
| 913.0 GB | 920.5 GB | 16-bit · bfloat — Brain floating point 16 — preferred for training | Runs on 8× H200 SXM 141 GB |
Selected BF16
Brain floating point 16916.5GBat 16K context
- Weights
- 912.2 GB
- KV cache
- 4.0 GB
- Overhead
- 0.3 GB
Server-class: ~916.5 GB at 16K context · about 8× H200 SXM 141 GB
Which Hardware Can Run MiniMax M1 40k?
BF16 · 913.0 GBMiniMax M1 40k (BF16) requires 913.0 GB of VRAM to load the model weights. For comfortable inference with headroom for KV cache and system overhead, 1187+ GB is recommended. Using the full 10240K context window can add up to 2516.1 GB, bringing total usage to 3429.1 GB. No consumer GPU has enough memory.
Needs server-class hardware
MiniMax M1 40k runs on a multi-GPU server · BF16 needs ~916.5 GB
| Hardware | Quant | Memory | Context | Hardware page | |
|---|---|---|---|---|---|
| 8× H200 SXM 141 GB · 1.1 TB | BF16 | 916.5 GB | 256K ctx | H200 SXM → | BF16 · 916.5 GB · 256K ctx |
| 6× B200 192 GB · 1.2 TB | BF16 | 916.5 GB | 256K ctx | B200 → | BF16 · 916.5 GB · 256K ctx |
By VRAM tier · typical GPUs and Macs at BF16
Consumer GPUs (up to 32 GB) and Macs up to 512 GB: too small for this model (needs ≥ 916.5 GB)8 GB tier →12 GB tier →16 GB tier →24 GB tier →Macs →
Alternatives to MiniMax M1 40k
MiniMax family · 10 models →Specifications
- Publisher
- MiniMax
- Family
- MiniMax
- Parameters
- 456.1B
- Active parameters
- 48.4B per token
- Architecture
- MiniMaxM1ForCausalLM
- Context Length
- 10,240,000 tokens
- Vocabulary Size
- 200,064
- Release Date
- 2025-06-05
- License
- Apache 2.0
Frequently Asked Questions
How much VRAM does MiniMax M1 40k need?
MiniMax M1 40k requires 913.0 GB of VRAM at BF16. Full 10240K context adds up to 2516.1 GB (3429.1 GB total).
VRAM = Weights + KV Cache + Overhead
Weights = 456.1B × 16 bits ÷ 8 = 912.2 GB
KV Cache + Overhead ≈ 0.8 GB (at 2K context + ~0.3 GB framework)
Fit ratings and hardware model lists check this model with room for a 16K-token context, which needs a little more memory.
KV Cache + Overhead ≈ 2516.9 GB (at full 10240K context)
VRAM usage by quantization
Can NVIDIA GeForce RTX 5090 run MiniMax M1 40k?
No — MiniMax M1 40k requires at least 913.0 GB at BF16, which exceeds the NVIDIA GeForce RTX 5090's 32 GB of VRAM.
Can I run MiniMax M1 40k on a Mac?
MiniMax M1 40k requires at least 913.0 GB at BF16. None of the 146 Macs we list has that much usable memory (the largest has 508.5 GB after macOS overhead).
Can I run MiniMax M1 40k locally?
Yes — MiniMax M1 40k can run locally on consumer hardware. Popular tools include Ollama, LM Studio, and llama.cpp.
What's the download size of MiniMax M1 40k?
At BF16, the download is about 912.18 GB.
Which GPUs can run MiniMax M1 40k?
No single consumer GPU has enough VRAM to run MiniMax M1 40k at BF16 (913.0 GB). Multi-GPU or professional hardware is required.
Which devices can run MiniMax M1 40k?
MiniMax M1 40k requires at least 913.0 GB at BF16, which exceeds the usable memory of every unified-memory device we list (the largest has 508.5 GB).