ToolSite MiniMax M2.7 VRAM requirements
MiniMax M2.7 has 228.7B parameters, of which about 10B are used per token; all 256 experts still have to be in memory. With an 8K-token context and one request it needs about 144 GB of GPU memory at Q4_K_M, 237 GB at FP8 and 471 GB at FP16/BF16. The published weights take 214 GB (FP8). It does not fit on a 24 GB card at Q4_K_M; the smallest single GPU that holds it with an 8K context is the 180 GB B200.
Open MiniMax M2.7 in the calculator
VRAM by quantization
Weights plus the KV cache for 8,192 tokens in FP16 and the runtime overhead (0.5 GB plus 10%). Each row opens the calculator with that setting. What the GGUF names mean.
| Precision | Weights | Total | Smallest setup |
|---|---|---|---|
| As published (FP8) | 214 GB | 238 GB | 2 × H200 |
| FP16 / BF16 | 426 GB | 471 GB | 4 × H200 |
| FP8 / INT8 | 213 GB | 237 GB | 2 × H200 |
| INT4 (AWQ / GPTQ) | 113 GB | 127 GB | H200 |
| GGUF Q8_0 | 226 GB | 252 GB | 2 × H200 |
| GGUF Q6_K | 175 GB | 195 GB | 2 × H200 |
| GGUF Q5_K_M | 151 GB | 169 GB | B200 |
| GGUF Q4_K_M | 129 GB | 144 GB | B200 |
| GGUF Q3_K_M | 104 GB | 117 GB | H200 |
| GGUF Q2_K | 89.2 GB | 101 GB | H200 |
KV cache at long context
All 62 layers use full attention. Each extra token of context adds 248 KB of FP16 cache per request. How the KV cache works.
| Context | KV cache, FP16 | KV cache, FP8 | Total at Q4_K_M |
|---|---|---|---|
| 4K tokens | 992 MB | 496 MB | 143 GB |
| 32K tokens | 7.75 GB | 3.88 GB | 151 GB |
| 128K tokens | 31.0 GB | 15.5 GB | 176 GB |
| 200K tokens | 48.4 GB | 24.2 GB | 196 GB |
Which GPUs can run MiniMax M2.7
With 8,192 tokens of context. Several GPUs means one tensor-parallel group of 2, 4 or 8 cards.
| GPU | Memory | Q4_K_M | FP8 |
|---|---|---|---|
| RTX 3060 | 12 GB | Needs more than 8 | Needs more than 8 |
| RTX 4060 Ti 16GB | 16 GB | Needs more than 8 | Needs more than 8 |
| RTX 3090 / 4090 | 24 GB | Needs 8 | Needs more than 8 |
| RTX 5090 | 32 GB | Needs 8 | Needs 8 |
| A100 40GB | 40 GB | Needs 4 | Needs 8 |
| Mac, 64 GB unified memory about 75% of it is usable by the GPU by default | 48 GB | Needs 4 | Needs 8 |
| L40S / RTX 6000 Ada | 48 GB | Needs 4 | Needs 8 |
| A100 / H100 80GB | 80 GB | Needs 2 | Needs 4 |
| Mac, 128 GB unified memory about 75% of it is usable by the GPU by default | 96 GB | Needs 2 | Needs 4 |
| H200 | 141 GB | Needs 2 | Needs 2 |
| B200 | 180 GB | Fits on one | Needs 2 |
How fast MiniMax M2.7 writes
Tokens per second for one request with 8,192 tokens of context, estimated from memory bandwidth and the parameters read per token. A dash means it does not fit on one card. Try other settings in the speed calculator.
| Hardware | Bandwidth | Q4_K_M | FP8 |
|---|---|---|---|
| RTX 3060 12GB | 360 GB/s | — | — |
| RTX 4090 | 1,008 GB/s | — | — |
| RTX 5090 | 1,792 GB/s | — | — |
| M4 Max Mac (128 GB) | 546 GB/s | — | — |
| M3 Ultra Mac Studio (512 GB) | 819 GB/s | 29–49 | 20–33 |
| H100 SXM | 3,350 GB/s | — | — |
| H200 | 4,800 GB/s | — | — |
Longest context on one GPU
How many tokens of context fit on a single card with one request and an FP16 KV cache. "Full" means the model's whole context window fits.
| GPU | Memory | Q4_K_M | Q8_0 | FP8 |
|---|---|---|---|---|
| RTX 3060 | 12 GB | No | No | No |
| RTX 4060 Ti 16GB | 16 GB | No | No | No |
| RTX 3090 / 4090 | 24 GB | No | No | No |
| RTX 5090 | 32 GB | No | No | No |
| A100 40GB | 40 GB | No | No | No |
| Mac, 64 GB unified memory | 48 GB | No | No | No |
| L40S / RTX 6000 Ada | 48 GB | No | No | No |
| A100 / H100 80GB | 80 GB | No | No | No |
| Mac, 128 GB unified memory | 96 GB | No | No | No |
| H200 | 141 GB | No | No | No |
| B200 | 180 GB | 141K | No | No |
Model details
- Parameters
- 228.7B (228,689,764,864)
- Experts
- 256 routed experts, all loaded
- Active per token
- 10B (estimated from the config)
- Layers
- All 62 layers use full attention
- Attention cache
- 8 KV heads × 128
- Context length
- 204,800 tokens
- Published weights
- 214 GB (FP8)
- On Hugging Face
- MiniMaxAI/MiniMax-M2.7
Other models
- DeepSeek V4 Flash 0731
- Qwen3.6 35B-A3B
- Qwen3.6 27B
- Qwen3.8 27B
- Qwen3.8 Flash Next
- Qwen3.8 2.4T-A95B
- Qwen3.5 9B
- Qwen3.5 122B-A10B
- Qwen3-Coder-Next
- DeepSeek V4.1 Flash
- DeepSeek V4 Flash
- DeepSeek V4 Pro
- DeepSeek V3.2
- GLM-5.3
- GLM-5.3 Flash
- GLM-5.2
- GLM-4.7 Flash
- Gemma 4 31B
- Gemma 4 26B-A4B
- Gemma 4 12B
- Gemma 4 E4B
- Kimi K3
- MiniMax M3
- MiMo V2.6 Flash
- MiMo V2.6 Pro
- Mistral Medium 3.5 128B
- Nemotron 3 Nano 4B
- Nemotron 3 Nano 30B-A3B
- Nemotron 3 Super 120B-A12B
- Xing 4.0 29B-A4B
- Muse Glimmer 30B
- MiniCPM5 2B
- Llama 3.1 8B
- Llama 3.1 70B
- Qwen3 8B
- Qwen3 30B-A3B
- gpt-oss-20b
- gpt-oss-120b
- DeepSeek V3 / R1
Numbers read from the model files on Hugging Face on .