MiniMax Models — Hardware Requirements
10 MiniMax models from MiniMax and the community, from the smallest that runs in 2.2 GB of VRAM at Q4 up to 456.4B parameters. Every row links to full quantization tables and GPU compatibility.
All MiniMax Models by Size
| Model | Params | Runs from (Q4) | Context | Publisher | Quant downloads |
|---|---|---|---|---|---|
| MiniMax M3 EAGLE3 | 3.3B | 2.7 GB | 1049K | ||
| MiniMax M2.5 | 228.7B | 139.6 GB | 197K | ||
| MiniMax M2.7 | 228.7B | 139.6 GB | 205K | ||
| MiniMax M2 | 228.7B | 139.6 GB | 197K | ||
| MiniMax M2.1 | 228.7B | 139.6 GB | 197K | ||
| MiniMax M3 | 427.0B | 258.0 GB | 1049K | ||
| MiniMax M3.1 Flash | 428B | 258.6 GB | 1049K | ||
| MiniMax VL 01 | 456.4B | 304.7 GB | 8K |
Community variants and fine-tunes(4)
| Model | Params | Runs from (Q4) | Context | Publisher | Quant downloads |
|---|---|---|---|---|---|
| MiniMax M3 EAGLE3 GQA | 3.1B | 2.2 GB | 1049K | ||
| MiniMax M2.7 BF16 Ultra Uncensored Heretic | 228.7B | 139.6 GB | 205K | ||
| MiniMax Text 01 | 456.1B | 309.0 GB | 10240K | ||
| MiniMax M1 80k | 456.1B | 309.0 GB | 10240K |
How MiniMax Compares — llmrun Composite Score
MiniMax M3 is the highest-scoring MiniMax model with an llmrun Composite Score of 112.6 — #27 among 106 open models. The top proprietary model, Gemini 4 Carbon, scores 158.6. Click a model to see its full benchmark breakdown.
Gemini 4 Carbon · proprietary158.6
GPT 6 Astra · proprietary156.2
Claude Opus 5.5 · proprietary153.0
GPT 6.1 Sol · proprietary151.7
GPT 6.1 Terra · proprietary150.1
Kimi K3.1 Pro142.8
MiMo V2.6 Pro RL133.3
Kimi K3132.7
MiMo V2.6 Flash RL131.9
GLM 5.3130.1
DeepSeek V4.1 Flash129.4
DeepSeek V4 Pro 0813129.2
MiniMax M3112.6
MiniMax M2.5111.8
MiniMax M2.7107.4
MiniMax M2.1106.2
MiniMax M2104.0
Frequently Asked Questions
- How much VRAM do I need to run a MiniMax model?
- The smallest MiniMax model, MiniMax M3 EAGLE3 GQA, runs from 2.2 GB of VRAM at Q4_K_M (smaller low-bit quantizations start at 1.6 GB with a noticeable quality loss). Larger family members need proportionally more — see the table above for every model.
- Which MiniMax models can I run on a 16 GB GPU?
- 2 of 12 MiniMax models fit in 16 GB of VRAM at Q4_K_M, including MiniMax M3 EAGLE3, MiniMax M3 EAGLE3 GQA.
- What is the most popular MiniMax model to run locally?
- MiniMax M2.5 is the most downloaded MiniMax model in local-friendly quantized formats. It runs from 139.6 GB of VRAM at Q4_K_M.
- How do MiniMax models score on benchmarks?
- MiniMax M3 leads the family with an llmrun Composite Score of 112.6, ranking #27 among 106 open models, while the top proprietary model, Gemini 4 Carbon, scores 158.6. See the comparison chart above for the full standings.