Total runs:
6.4M
Run Growth:
-285.4K
Growth Rate:
-4.29%
Sponsored by
APIMart
unsloth / MiniMax-M3-GGUF
Model's Last Updated: October 12 2026
image-text-to-text
Introduction of MiniMax-M3-GGUF
Model Details of MiniMax-M3-GGUF
Read our How to Run MiniMax M3 Guide!
See Unsloth Dynamic 2.0 GGUFs for our quantization benchmarks.
- EXPERIMENTAL GGUF / support for MiniMax-M3
- Jun 12 Update: You can now run MiniMax M3 in Unsloth Studio. See our Guide .
- Example of MiniMax M3 (5-bit GGUF) running in Unsloth Studio:
EXPERIMENTAL GGUF / support for MiniMax-M3 in llama.cpp:
MiniMax-M3 support in llama.cpp is preliminary and not yet in a released build. To run these GGUFs, build llama.cpp from [PR #24523](https://github.com/ggml-org/llama.cpp/pull/24523):
git clone https://github.com/ggml-org/llama.cpp
cd llama.cpp
git fetch origin pull/24523/head:minimax-m3
git checkout minimax-m3
cmake -B build -DGGML_CUDA=ON
cmake --build build --config Release -j --target llama-cli llama-server
Then run a quant. The model is large (~428B params), so offload across GPUs with
-ngl 99
or keep the weights in CPU RAM:
./build/bin/llama-cli -hf unsloth/MiniMax-M3-GGUF:UD-IQ1_M
Note: MiniMax Sparse Attention is not supported yet, so inference falls back to dense attention.
MiniMax-M3
Highlights:
- Native Multimodality: M3 undergoes mixed-modality training from the very first step, enabling deeper semantic fusion across text, image, and video.
- Context Scaling via Sparse Attention: M3 introduces MiniMax Sparse Attention (MSA) to improve long context efficiency. M3 delivers 9× prefill and 15× decode speedups compared to M2 at 1M context, reducing per-token compute to 1/20.
- Coding & Cowork Capability: M3 achieves frontier-level performance across long-horizon agentic benchmarks, excelling in both coding and cowork.
Model Details
| Architecture | MoE + MSA (MiniMax Sparse Attention) |
| Total Parameters | ~428B |
| Activated Parameters | ~23B |
| Experts | 128 (4 active per token) |
| Layers | 60 |
| Context Length | 1M tokens |
| Modalities | Text, Image, Video |
| Precision | bfloat16 |
| Transformers |
≥ 4.52.4 (
trust_remote_code=True
)
|
| License | MiniMax Community License |
How to Use
M3 supports two reasoning modes:
- thinking — for complex reasoning, agentic tasks, and long-horizon collaboration.
- non-thinking — for latency-sensitive scenarios such as chat and code completion.
Local Deployment
Download the model:
hf download MiniMaxAI/MiniMax-M3 --local-dir MiniMax-M3
We recommend the following inference frameworks (listed alphabetically) to serve the model:
SGLang
We recommend using SGLang to serve MiniMax-M3. Please refer to our SGLang Deployment Guide .
vLLM
We recommend using vLLM to serve MiniMax-M3. Please refer to our vLLM Deployment Guide .
Transformers
We recommend using Transformers to serve MiniMax-M3. Please refer to our Transformers Deployment Guide .
ModelScope
You can also get model weights from ModelScope .
Inference Parameters
We recommend the following parameters for best performance:
temperature=1.0
,
top_p=0.95
,
top_k=40
. Default system prompt:
You are a helpful assistant. Your name is MiniMax-M3 and was built by MiniMax.
Tool Calling Guide
Please refer to our Tool Calling Guide .
Contact Us
Contact us at [email protected] .
Runs of unsloth MiniMax-M3-GGUF on huggingface.co
6.3K
Total runs
0
24-hour runs
-215
3-day runs
-81
7-day runs
-157
30-day runs
More Information About MiniMax-M3-GGUF huggingface.co Model
Url of MiniMax-M3-GGUF
MiniMax-M3-GGUF huggingface.co Url
Provider of MiniMax-M3-GGUF huggingface.co
Other API from unsloth
Updated:August 20 2026
Total runs:
2.0M
Run Growth:
977.9K
Growth Rate:
49.96%
Total runs:
1.6M
Run Growth:
719.2K
Growth Rate:
44.47%
Total runs:
1.5M
Run Growth:
1.4M
Growth Rate:
100.00%
Total runs:
1.3M
Run Growth:
-217.4K
Growth Rate:
-16.74%
Total runs:
1.3M
Run Growth:
-28.0K
Growth Rate:
-2.27%
Total runs:
1.3M
Run Growth:
337.9K
Growth Rate:
27.14%
Total runs:
1.1M
Run Growth:
-11.2M
Growth Rate:
-758.88%
Total runs:
1.1M
Run Growth:
36.7K
Growth Rate:
3.38%
Total runs:
1.1M
Run Growth:
-455.6K
Growth Rate:
-40.49%
Total runs:
1.1M
Run Growth:
720.6K
Growth Rate:
66.92%
Total runs:
959.5K
Run Growth:
784.8K
Growth Rate:
82.09%
Total runs:
934.7K
Run Growth:
281.7K
Growth Rate:
29.89%
Total runs:
932.5K
Run Growth:
767.7K
Growth Rate:
82.32%
Total runs:
919.6K
Run Growth:
302.8K
Growth Rate:
33.47%
Total runs:
861.5K
Run Growth:
0
Growth Rate:
0.00%
Total runs:
750.1K
Run Growth:
-633.4K
Growth Rate:
-83.55%
Total runs:
707.0K
Run Growth:
125.4K
Growth Rate:
17.74%
Total runs:
632.2K
Run Growth:
249.5K
Growth Rate:
40.14%
Total runs:
610.6K
Run Growth:
112.7K
Growth Rate:
18.57%
Total runs:
536.1K
Run Growth:
-9.9K
Growth Rate:
-1.86%
Total runs:
532.3K
Run Growth:
171.2K
Growth Rate:
35.06%
Total runs:
479.7K
Run Growth:
163.5K
Growth Rate:
34.08%
Total runs:
451.8K
Run Growth:
-648.4K
Growth Rate:
-143.52%
Total runs:
424.5K
Run Growth:
391.6K
Growth Rate:
99.95%
Total runs:
422.8K
Run Growth:
676
Growth Rate:
0.16%
Total runs:
422.3K
Run Growth:
-81.5K
Growth Rate:
-19.59%
Total runs:
410.4K
Run Growth:
-92.8K
Growth Rate:
-22.42%
Total runs:
401.6K
Run Growth:
97.6K
Growth Rate:
24.31%
Total runs:
387.0K
Run Growth:
-267.2K
Growth Rate:
-69.03%
Total runs:
379.3K
Run Growth:
125.6K
Growth Rate:
33.14%
Total runs:
378.9K
Run Growth:
101.9K
Growth Rate:
27.17%
Total runs:
373.6K
Run Growth:
-4.0K
Growth Rate:
-1.08%
Total runs:
367.1K
Run Growth:
47.1K
Growth Rate:
12.83%
Total runs:
362.4K
Run Growth:
330.3K
Growth Rate:
100.00%
Total runs:
350.2K
Run Growth:
-9.5K
Growth Rate:
-2.71%
Total runs:
342.7K
Run Growth:
174.0K
Growth Rate:
53.14%
Total runs:
330.5K
Run Growth:
108.7K
Growth Rate:
32.88%
Total runs:
318.4K
Run Growth:
283.5K
Growth Rate:
89.03%
Total runs:
315.9K
Run Growth:
207.2K
Growth Rate:
66.82%
Total runs:
313.5K
Run Growth:
-49.3K
Growth Rate:
-15.73%
Total runs:
306.2K
Run Growth:
-27.4K
Growth Rate:
-8.94%
Total runs:
302.7K
Run Growth:
-177.8K
Growth Rate:
-58.14%
Total runs:
294.0K
Run Growth:
-137.2K
Growth Rate:
-46.67%
Total runs:
287.0K
Run Growth:
-328.9K
Growth Rate:
-108.40%
Total runs:
280.0K
Run Growth:
-118.9K
Growth Rate:
-43.08%
Total runs:
277.3K
Run Growth:
-660.1K
Growth Rate:
-223.12%
Total runs:
253.4K
Run Growth:
-129.9K
Growth Rate:
-51.27%
Total runs:
246.8K
Run Growth:
26.5K
Growth Rate:
10.86%
Total runs:
243.2K
Run Growth:
224.4K
Growth Rate:
91.95%
Total runs:
243.2K
Run Growth:
-58.2K
Growth Rate:
-28.52%
Total runs:
232.8K
Run Growth:
-106.2K
Growth Rate:
-45.45%
Total runs:
230.9K
Run Growth:
43.5K
Growth Rate:
24.91%
Total runs:
223.2K
Run Growth:
65.7K
Growth Rate:
29.41%
Total runs:
218.1K
Run Growth:
146.7K
Growth Rate:
70.01%
Total runs:
214.2K
Run Growth:
-49.1K
Growth Rate:
-22.94%
Total runs:
211.6K
Run Growth:
-129.5K
Growth Rate:
-58.26%
Total runs:
207.6K
Run Growth:
89.5K
Growth Rate:
43.54%
Total runs:
196.4K
Run Growth:
-415.5K
Growth Rate:
-192.79%
Total runs:
190.9K
Run Growth:
22.2K
Growth Rate:
11.64%
Total runs:
189.4K
Run Growth:
-3.8K
Growth Rate:
-2.00%
Total runs:
180.3K
Run Growth:
172.9K
Growth Rate:
95.91%
Total runs:
179.7K
Run Growth:
-223.2K
Growth Rate:
-124.21%
Total runs:
176.3K
Run Growth:
115.7K
Growth Rate:
65.63%
Total runs:
175.6K
Run Growth:
-391.8K
Growth Rate:
-248.36%
Total runs:
173.5K
Run Growth:
13.1K
Growth Rate:
7.53%
Total runs:
163.9K
Run Growth:
-22.7K
Growth Rate:
-13.82%
Total runs:
162.5K
Run Growth:
148.1K
Growth Rate:
99.55%
Total runs:
161.4K
Run Growth:
33.7K
Growth Rate:
20.90%
Total runs:
154.7K
Run Growth:
-984.3K
Growth Rate:
-635.35%
Total runs:
151.5K
Run Growth:
140.0K
Growth Rate:
92.42%
Total runs:
148.9K
Run Growth:
-440.4K
Growth Rate:
-295.78%
Total runs:
148.8K
Run Growth:
147.0K
Growth Rate:
98.82%
Total runs:
148.6K
Run Growth:
50.0K
Growth Rate:
33.62%
Total runs:
146.6K
Run Growth:
-36.5K
Growth Rate:
-25.38%
Total runs:
141.0K
Run Growth:
-156.8K
Growth Rate:
-111.23%
Total runs:
138.9K
Run Growth:
96.1K
Growth Rate:
69.18%
Total runs:
133.1K
Run Growth:
18.0K
Growth Rate:
13.53%
Total runs:
131.7K
Run Growth:
39.1K
Growth Rate:
29.69%
Total runs:
130.9K
Run Growth:
64.6K
Growth Rate:
49.46%
Total runs:
124.0K
Run Growth:
119.8K
Growth Rate:
96.59%
Total runs:
122.0K
Run Growth:
64.0K
Growth Rate:
53.47%
Total runs:
121.1K
Run Growth:
-454.1K
Growth Rate:
-330.61%
Total runs:
117.4K
Run Growth:
-108.9K
Growth Rate:
-93.32%
Total runs:
115.3K
Run Growth:
22.2K
Growth Rate:
19.30%
Total runs:
114.9K
Run Growth:
2.5K
Growth Rate:
2.18%
Total runs:
114.6K
Run Growth:
100.2K
Growth Rate:
89.73%
Total runs:
111.7K
Run Growth:
-119.1K
Growth Rate:
-113.36%
Total runs:
110.4K
Run Growth:
-8.1K
Growth Rate:
-7.40%
Total runs:
105.9K
Run Growth:
-244.2K
Growth Rate:
-230.67%
Total runs:
105.4K
Run Growth:
3.1K
Growth Rate:
2.94%
Total runs:
104.4K
Run Growth:
20.2K
Growth Rate:
19.40%
Total runs:
94.4K
Run Growth:
-60.4K
Growth Rate:
-63.94%