CanItRun Logocanitrun.

MiniMax M1 456B

MiniMax M1 456B needs roughly 258.4GB VRAM at Q4_K_M quantization (1024.5GB at FP16). 5 GPUs we track can run it fully in VRAM at 8k context.

MiniMax456B params46B active (MoE)1024k contextApache 2.0Commercial use ok

VRAM at each quantization

Assumes 8k context. KV cache grows linearly with context length.

QuantWeightsKV cacheTotal
FP16912.0 GB2.68 GB1024.5 GB
Q8456.0 GB2.68 GB513.7 GB
Q6_K342.0 GB2.68 GB386.1 GB
Q5_K_M285.0 GB2.68 GB322.2 GB
Q4_K_M228.0 GB2.68 GB258.4 GB
Q3_K_M182.4 GB2.68 GB207.3 GB
Q2_K136.8 GB2.68 GB156.2 GB

Benchmarks

GPUs that run MiniMax M1 456B natively (5)

Notes

Hybrid lightning attention (linear+softmax) reasoning model with 1M context. 40k and 80k thinking-budget variants available.

Hugging Face ↗Released 2025-06-16