EmbeddedLLM / MiniMax-M3-FP8-dynamic

huggingface.co
Total runs: 7.5K
24-hour runs: -273
7-day runs: -3.1K
30-day runs: -22.9K
Model's Last Updated: June 27 2026
image-text-to-text

Introduction of MiniMax-M3-FP8-dynamic

Model Details of MiniMax-M3-FP8-dynamic

MiniMax-M3-FP8-dynamic

Model Overview

This model is an FP8 dynamic quantized version of MiniMaxAI/MiniMax-M3 .

  • Base model: MiniMaxAI/MiniMax-M3
  • Optimization: FP8 dynamic quantization
  • Format: safetensors / compressed-tensors
  • Validated runtime: vLLM OpenAI-compatible server
  • Tested hardware: AMD MI350, tensor parallel size 8

MiniMax-M3 is a native multimodal MoE model. The original model card describes it as a ~428B parameter model with ~23B activated parameters and 1M context support.

License

This quantized checkpoint follows the license terms of the base model, MiniMaxAI/MiniMax-M3 . The Hugging Face model-card metadata uses license: other because the MiniMax community license is not one of the Hub's enumerated license identifiers.

Model Optimizations

This checkpoint uses FP8 dynamic quantization to reduce memory and disk requirements while preserving model quality. Validation below compares this quantized checkpoint against the BF16 MiniMaxAI/MiniMax-M3 baseline.

Evaluation

The model was evaluated against BF16 MiniMaxAI/MiniMax-M3 . Scores are averaged across seeds.

Benchmark MiniMaxAI/MiniMax-M3 EmbeddedLLM/MiniMax-M3-FP8-dynamic Recovery (%)
GSM8k Platinum 95.81 95.92 100.12
IfEval 80.65 79.42 98.47
AIME 2025 20.83 19.17 92.00
GPQA diamond 77.78 77.95 100.22
Math 500 81.20 79.93 98.44
Lcb Codegeneration V6 37.14 35.62 95.90
MMLU Pro Chat 79.85 79.62 99.72
Evaluation Setup
  • Standard seeds: 42, 1234, 4158
  • AIME 2025 seeds: 42, 1234, 4158, 5322, 1356, 9843, 3344, 5678
  • GSM8K Platinum cap: max_gen_toks=64000
  • IFEval, AIME, GPQA, Math 500, MMLU Pro Chat cap: max_gen_toks=4096
  • LiveCodeBench v6 cap: max_gen_toks=2048
  • MiniMax thinking mode: disabled
  • Runners: lm-eval harness and lighteval through LiteLLM endpoint mode

Runs of EmbeddedLLM MiniMax-M3-FP8-dynamic on huggingface.co

7.5K
Total runs
-273
24-hour runs
-700
3-day runs
-3.1K
7-day runs
-22.9K
30-day runs

More Information About MiniMax-M3-FP8-dynamic huggingface.co Model

More MiniMax-M3-FP8-dynamic license Visit here:

https://choosealicense.com/licenses/other

MiniMax-M3-FP8-dynamic huggingface.co

MiniMax-M3-FP8-dynamic huggingface.co is an AI model on huggingface.co that provides MiniMax-M3-FP8-dynamic's model effect (), which can be used instantly with this EmbeddedLLM MiniMax-M3-FP8-dynamic model. huggingface.co supports a free trial of the MiniMax-M3-FP8-dynamic model, and also provides paid use of the MiniMax-M3-FP8-dynamic. Support call MiniMax-M3-FP8-dynamic model through api, including Node.js, Python, http.

MiniMax-M3-FP8-dynamic huggingface.co Url

https://huggingface.co/EmbeddedLLM/MiniMax-M3-FP8-dynamic

EmbeddedLLM MiniMax-M3-FP8-dynamic online free

MiniMax-M3-FP8-dynamic huggingface.co is an online trial and call api platform, which integrates MiniMax-M3-FP8-dynamic's modeling effects, including api services, and provides a free online trial of MiniMax-M3-FP8-dynamic, you can try MiniMax-M3-FP8-dynamic online for free by clicking the link below.

EmbeddedLLM MiniMax-M3-FP8-dynamic online free url in huggingface.co:

https://huggingface.co/EmbeddedLLM/MiniMax-M3-FP8-dynamic

MiniMax-M3-FP8-dynamic install

MiniMax-M3-FP8-dynamic is an open source model from GitHub that offers a free installation service, and any user can find MiniMax-M3-FP8-dynamic on GitHub to install. At the same time, huggingface.co provides the effect of MiniMax-M3-FP8-dynamic install, users can directly use MiniMax-M3-FP8-dynamic installed effect in huggingface.co for debugging and trial. It also supports api for free installation.

MiniMax-M3-FP8-dynamic install url in huggingface.co:

https://huggingface.co/EmbeddedLLM/MiniMax-M3-FP8-dynamic

Url of MiniMax-M3-FP8-dynamic

MiniMax-M3-FP8-dynamic huggingface.co Url

Provider of MiniMax-M3-FP8-dynamic huggingface.co

EmbeddedLLM
ORGANIZATIONS

Other API from EmbeddedLLM