MiniMax
MiniMax M2.7 (230B parameters) requires approximately 145.6 GB of VRAM with UD-IQ4_XS quantization. As a Mixture of Experts model with 10B active parameters, it uses less memory than its total parameter count suggests. For the best balance of quality and speed, we recommend hardware with at least 168 GB of VRAM.
Get started
— copy & paste to run locallyCopy-paste commands to run MiniMax M2.7 on your machine.
Run
lms load MiniMax-M2.7 && lms server startQuick specs
About this model
Inference speed
Estimated decode speed (tokens/sec) for MiniMax M2.7 at UD-IQ4_XS across popular GPUs and Apple Silicon, including multi-GPU rigs, using the fastest local runtime per device. Fastest is Mac Studio M3 Ultra 256GB at ~20 tok/s. Speed is memory-bandwidth bound, so cards that fit the whole model in VRAM run far faster than ones that offload to system RAM.
| GPU / Mac | Memory | Quant | Speed (tok/s) | Fits? |
|---|---|---|---|---|
Mac Studio M3 Ultra 256GB | 256 GB | UD-IQ4_XS | 20.0 | Tight |
Mac Studio M2 Ultra 128GB | 128 GB | UD-IQ4_XS | 8.3 | Too big |
Mac Studio M1 Ultra 128GB | 128 GB | UD-IQ4_XS | 7.9 | Too big |
MacBook Pro M4 Max 128GB | 128 GB | UD-IQ4_XS | 6.2 | Too big |
MacBook Pro M4 Max 64GB | 64 GB | UD-IQ4_XS | 5.6 | Too big |
 RTX 5090 32GB | 32 GB | UD-IQ4_XS | 4.2 | Too big |
2× RX 7900 XTX 24GB | 48 GB | UD-IQ4_XS | 4.1 | Too big |
MacBook Pro M3 Max 64GB | 64 GB | UD-IQ4_XS | 3.9 | Too big |
MacBook Pro M1 Max 64GB | 64 GB | UD-IQ4_XS | 3.6 | Too big |
MacBook Pro M4 Pro 48GB | 48 GB | UD-IQ4_XS | 3.4 | Too big |
| 2× RTX 4090 24GB | 48 GB | UD-IQ4_XS | 2.8 | Too big |
| RTX 4090 24GB | 24 GB | UD-IQ4_XS | 2.7 | Too big |
RX 7900 XTX 24GB | 24 GB | UD-IQ4_XS | 2.4 | Too big |
| 2× RTX 3090 24GB | 48 GB | UD-IQ4_XS | 2.4 | Too big |
| RTX 3090 24GB | 24 GB | UD-IQ4_XS | 2.3 | Too big |
| RTX 4080 Super 16GB | 16 GB | UD-IQ4_XS | 2.1 | Too big |
| 4× RTX 3060 12GB | 48 GB | UD-IQ4_XS | 2.1 | Too big |
| RTX 4070 12GB | 12 GB | UD-IQ4_XS | 2.0 | Too big |
| RTX 3060 12GB | 12 GB | UD-IQ4_XS | 2.0 | Too big |
| RTX 4060 8GB | 8 GB | UD-IQ4_XS | 2.0 | Too big |
Estimates for single-stream decoding at UD-IQ4_XS; real tokens/sec varies with prompt length, context, batch size, and runtime build. Prompt processing (prefill) is faster than the decode figures shown here.
Quick picks
Best hardware
Run this model
Quantization
How much VRAM MiniMax M2.7 (230B) needs at each GGUF quant, and whether it fits a 24 GB card (RTX 4090 / 3090).
| Quant | Bits | VRAM (weights) | Quality | Fits 24 GB? |
|---|---|---|---|---|
| Q2_K | 2 | 89.7 GB | Low | Too big |
| Q3_K_S | 3 | 112.7 GB | Low | Too big |
| NVFP4 | 4 | 128.8 GB | Medium | Too big |
| Q4_K_M | 4 | 140.3 GB | Medium | Too big |
| Q5_K_M | 5 | 165.6 GB | High | Too big |
| Q6_K | 6 | 188.6 GB | High | Too big |
| Q8_0 | 8 | 246.1 GB | Very High | Too big |
| F16 | 16 | 471.5 GB | Maximum | Too big |
VRAM shown is quantized weights only; add ~1–3 GB runtime overhead plus KV cache for your context length. Lower quants trade quality for memory — Q4_K_M is the usual sweet spot; Q2/Q3 only when you must fit a bigger model.
Hardware compatibility
Computing compatibility...
Memory breakdown
Frequently asked questions
MiniMax M2.7 (230B parameters) requires approximately 145.6 GB of VRAM with UD-IQ4_XS quantization. Lower quantizations like Q4_K_M use less memory but may reduce quality.
Yes, Mac Studio M3 Ultra 256GB can run MiniMax M2.7 with a compatibility score of 85/100. It provides 256 GB of memory and achieves approximately 20.0 tokens per second.
The recommended quantization for MiniMax M2.7 is UD-IQ4_XS, which offers the best balance between model quality and memory efficiency. Higher quantizations preserve more quality but require more VRAM.
The top recommended hardware for MiniMax M2.7: AMD Instinct MI325X 256GB (score: 93/100), AMD Instinct MI350X 288GB (score: 91/100), NVIDIA B200 180GB (score: 90/100). These provide the best combination of memory, bandwidth, and compute for running this model locally.
Yes, MiniMax M2.7 is well-suited for coding as well as agentic, chat. It was designed with these use cases in mind.
See also