MiniMax M3
417.1B total, 23B active · 1024k context · minimax-community · released 2026-06-02
chatcodingagenticvisionreasoning
In short
MiniMax M3 is an open model from MiniMax with 417.1B parameters (23B active per token). At mlx-4bit, MiniMax M3 needs about 227 GB of memory at 8k context and 233 GB at 64k, counting the weights, the KV cache and runtime buffers. Nothing we price in Germany runs MiniMax M3 fully in memory at 64k context at a usable speed; see the table for what comes closest.
Updated . Speeds are estimates; how they are worked out.
What to run it on in Germany
Fully in memory at 64k context, priced at retailers today.
Something we price holds it fully in memory, but none reaches 15 tok/s, our bar for recommending it. The table below has the speeds.
All hardware that runs it
Everything we price, best value first. Two cards split the model by layers.
| Hardware | Quant | Fits | Decode | Price | tok/s per 1,000 |
|---|---|---|---|---|---|
| Mac Studio M3 Ultra 512GB | UD-Q6_K | fits | 12 tok/s | 18.995 € | 0.6 |
39 other cards and machines we price cannot hold it at this context.
Memory needed
Weights, KV cache and runtime buffers at each context length (KV cache at fp16).
| Quant | File | @8k | @32k | @64k | @128k |
|---|---|---|---|---|---|
| mlx-4bitmlx | 224.9 GB | 226.6 GB | 229.5 GB | 233.5 GB | 241.3 GB |
| UD-Q4_K_Mgguf | 246.7 GB | 248.4 GB | 251.4 GB | 255.3 GB | 263.2 GB |
| UD-Q5_K_Mgguf | 296.8 GB | 298.5 GB | 301.5 GB | 305.4 GB | 313.3 GB |
| UD-Q6_Kgguf | 330.9 GB | 332.6 GB | 335.5 GB | 339.5 GB | 347.4 GB |
| Q8_0gguf | 422.5 GB | 424.1 GB | 427.1 GB | 431.0 GB | 438.9 GB |
Run it
Copy a command for the runtime you use.
- mlx-4bit
- LM Studio
lms get mlx-community/MiniMax-M3-4bit - MLX
mlx_lm.server --model mlx-community/MiniMax-M3-4bit
- LM Studio
- UD-Q4_K_M
- Ollama
ollama run hf.co/unsloth/MiniMax-M3-GGUF:UD-Q4_K_M - LM Studio
lms get unsloth/MiniMax-M3-GGUF@ud-q4_k_m - llama.cpp
llama-server -hf unsloth/MiniMax-M3-GGUF:UD-Q4_K_M
- Ollama
- UD-Q5_K_M
- Ollama
ollama run hf.co/unsloth/MiniMax-M3-GGUF:UD-Q5_K_M - LM Studio
lms get unsloth/MiniMax-M3-GGUF@ud-q5_k_m - llama.cpp
llama-server -hf unsloth/MiniMax-M3-GGUF:UD-Q5_K_M
- Ollama
- UD-Q6_K
- Ollama
ollama run hf.co/unsloth/MiniMax-M3-GGUF:UD-Q6_K - LM Studio
lms get unsloth/MiniMax-M3-GGUF@ud-q6_k - llama.cpp
llama-server -hf unsloth/MiniMax-M3-GGUF:UD-Q6_K
- Ollama
- Q8_0
- Ollama
ollama run hf.co/unsloth/MiniMax-M3-GGUF:Q8_0 - LM Studio
lms get unsloth/MiniMax-M3-GGUF@q8_0 - llama.cpp
llama-server -hf unsloth/MiniMax-M3-GGUF:Q8_0
- Ollama
Notes
Custom "minimax-community" licence (license: other); the card doesn't state commercial-use terms, so check the LICENSE file. Supports three thinking modes: enabled, adaptive, disabled. Sparse or compressed attention: KV read per token is estimated as full attention. Excludes 9.9B parameters of multi-token-prediction layers, which local runtimes do not load. Active parameters as the card states them (23.0B); the config gives 16.9B.
Questions
- How much VRAM does MiniMax M3 need?
- At mlx-4bit, MiniMax M3 needs about 227 GB of memory at 8k context and 233 GB at 64k, counting the weights, the KV cache and runtime buffers.
- Can I run MiniMax M3 on a 16 GB GPU?
- No: it does not fit in 16 GB at 64k context, even with system RAM.
- Can I run MiniMax M3 on a 24 GB GPU?
- No: it does not fit in 24 GB at 64k context, even with system RAM.
- How do I run MiniMax M3 locally?
- With Ollama: ollama run hf.co/unsloth/MiniMax-M3-GGUF:UD-Q4_K_M (the UD-Q4_K_M build). LM Studio and llama.cpp commands are on this page.
- What licence is MiniMax M3 released under?
- minimax-community.