How much GPU memory does MiniMax-M2.7-NVFP4 need?

About 160 GB for conversations of up to 8,192 tokens: 139.89 GB of weights, the cache for one full-length conversation and the runtime's own memory.

What is the cheapest GPU to run MiniMax-M2.7-NVFP4?

Right now 2x RTX PRO 6000D at $2.12 an hour. Prices are live offers from GPU clouds, checked every 15 minutes.

Can MiniMax-M2.7-NVFP4 run on an RTX 4090?

Not on one: it needs about 160 GB and an RTX 4090 has 24 GB. Use a bigger GPU, or a smaller (quantized) version of the model.

How do I deploy MiniMax-M2.7-NVFP4 as an API?

Pick a machine on its GridPup page and press Deploy: GridPup rents the GPU, downloads the model from Hugging Face and serves it at an https address with keys and OpenAI's chat API, billed by the hour.

Models / MiniMax-M2.7 / MiniMax-M2.7 ~229B

nvidia/MiniMax-M2.7-NVFP4

Text generation · served with vLLM · ↓ 5.1K downloads a month · View on Hugging Face

nvidia/MiniMax-M2.7-NVFP4 needs about 160 GB of GPU memory for conversations of up to 8,192 tokens. It runs on 2x RTX PRO 6000D from $2.12/h right now.

Weights: 140 GB
GPU memory needed: about 160 GB
Conversation length: 8,192 tokens
Model allows: 196,608 tokens
Memory · disk: 77.9 GB · 194 GB
Architecture: MiniMaxM2ForCausalLM
Licence: other

Versions of MiniMax-M2.7 ~229B

The same model packaged differently: a quantized version needs less GPU memory and runs on a cheaper machine.

Version	Runtime	Weights	GPU memory	Cheapest machine now
OriginalMiniMaxAI/MiniMax-M2.7	vLLM	230 GB	260.2 GB	$6.39/h 4x RTX PRO 6000
NVFP4 (nvidia)nvidia/MiniMax-M2.7-NVFP4	vLLM	140 GB	160 GB	$2.07/h 2x RTX PRO 6000D

Hardware and deploy

Endpoint name

$2.12/h from your balance while it runs, billed by the hour; stop it from its page. GridPup rents the machine, which downloads the model from Hugging Face (usually ready in 10 to 15 minutes), and gives you an https address, keys and a Playground.

Questions

How much GPU memory does MiniMax-M2.7-NVFP4 need?: About 160 GB for conversations of up to 8,192 tokens: 139.89 GB of weights, the cache for one full-length conversation and the runtime's own memory.
What is the cheapest GPU to run MiniMax-M2.7-NVFP4?: Right now 2x RTX PRO 6000D at $2.12 an hour. Prices are live offers from GPU clouds, checked every 15 minutes.
Can MiniMax-M2.7-NVFP4 run on an RTX 4090?: Not on one: it needs about 160 GB and an RTX 4090 has 24 GB. Use a bigger GPU, or a smaller (quantized) version of the model.
How do I deploy MiniMax-M2.7-NVFP4 as an API?: Pick a machine on its GridPup page and press Deploy: GridPup rents the GPU, downloads the model from Hugging Face and serves it at an https address with keys and OpenAI's chat API, billed by the hour.

Loading…

nvidia/MiniMax-M2.7-NVFP4

Text generation · served with vLLM · ↓ 5.1K downloads a month · View on Hugging Face

nvidia/MiniMax-M2.7-NVFP4 needs about 160 GB of GPU memory for conversations of up to 8,192 tokens. It runs on 2x RTX PRO 6000D from $2.12/h right now.

Weights

140 GB

GPU memory needed

about 160 GB

Conversation length

8,192 tokens

Model allows

196,608 tokens

Memory · disk

77.9 GB · 194 GB

Architecture

MiniMaxM2ForCausalLM

Licence

other

Versions of MiniMax-M2.7 ~229B

The same model packaged differently: a quantized version needs less GPU memory and runs on a cheaper machine.

Version	Runtime	Weights	GPU memory	Cheapest machine now
OriginalMiniMaxAI/MiniMax-M2.7	vLLM	230 GB	260.2 GB	$6.39/h 4x RTX PRO 6000
NVFP4 (nvidia)nvidia/MiniMax-M2.7-NVFP4	vLLM	140 GB	160 GB	$2.07/h 2x RTX PRO 6000D

Hardware and deploy

Endpoint name

Questions

How much GPU memory does MiniMax-M2.7-NVFP4 need?: About 160 GB for conversations of up to 8,192 tokens: 139.89 GB of weights, the cache for one full-length conversation and the runtime's own memory.
What is the cheapest GPU to run MiniMax-M2.7-NVFP4?: Right now 2x RTX PRO 6000D at $2.12 an hour. Prices are live offers from GPU clouds, checked every 15 minutes.
Can MiniMax-M2.7-NVFP4 run on an RTX 4090?: Not on one: it needs about 160 GB and an RTX 4090 has 24 GB. Use a bigger GPU, or a smaller (quantized) version of the model.
How do I deploy MiniMax-M2.7-NVFP4 as an API?: Pick a machine on its GridPup page and press Deploy: GridPup rents the GPU, downloads the model from Hugging Face and serves it at an https address with keys and OpenAI's chat API, billed by the hour.