Can an open supply MoE really energy agentic coding workflows at a fraction of flagship mannequin prices whereas sustaining long-horizon device use throughout MCP, shell, browser, retrieval, and code? MiniMax group has simply launched MiniMax-M2, a combination of specialists MoE mannequin optimized for coding and agent workflows. The weights are printed on Hugging Face underneath the MIT license, and the mannequin is positioned as for finish to finish device use, multi file modifying, and lengthy horizon plans, It lists 229B complete parameters with about 10B lively per token, which retains reminiscence and latency in verify throughout agent loops.
Structure and why activation dimension issues?
MiniMax-M2 is a compact MoE that routes to about 10B lively parameters per token. The smaller activations scale back reminiscence stress and tail latency in plan, act, and confirm loops, and permit extra concurrent runs in CI, browse, and retrieval chains. That is the efficiency price range that permits the pace and value claims relative to dense fashions of comparable high quality.
MiniMax-M2 is an interleaved considering mannequin. The analysis group wrapped inside reasoning in
Benchmarks that focus on coding and brokers
The MiniMax group experiences a set of agent and code evaluations are nearer to developer workflows than static QA. On Terminal Bench, the desk exhibits 46.3. On Multi SWE Bench, it exhibits 36.2. On BrowseComp, it exhibits 44.0. SWE Bench Verified is listed at 69.4 with the scaffold element, OpenHands with 128k context and 100 steps.
MiniMax’s official announcement stresses 8% of Claude Sonnet pricing, and close to 2x pace, plus a free entry window. The identical notice supplies the particular token costs and the trial deadline.
Comparability M1 vs M2
| Side | MiniMax M1 | MiniMax M2 |
|---|---|---|
| Complete parameters | 456B complete | 229B in mannequin card metadata, mannequin card textual content says 230B complete |
| Energetic parameters per token | 45.9B lively | 10B lively |
| Core design | Hybrid Combination of Specialists with Lightning Consideration | Sparse Combination of Specialists focusing on coding and agent workflows |
| Pondering format | Pondering price range variants 40k and 80k in RL coaching, no assume tag protocol required | Interleaved considering with |
| Benchmarks highlighted | AIME, LiveCodeBench, SWE-bench Verified, TAU-bench, lengthy context MRCR, MMLU-Professional | Terminal-Bench, Multi SWE-Bench, SWE-bench Verified, BrowseComp, GAIA textual content solely, Synthetic Evaluation intelligence suite |
| Inference defaults | temperature 1.0, prime p 0.95 | mannequin card exhibits temperature 1.0, prime p 0.95, prime ok 40, launch web page exhibits prime ok 20 |
| Serving steering | vLLM beneficial, Transformers path additionally documented | vLLM and SGLang beneficial, device calling information offered |
| Major focus | Lengthy context reasoning, environment friendly scaling of take a look at time compute, CISPO reinforcement studying | Agent and code native workflows throughout shell, browser, retrieval, and code runners |
Key Takeaways
- M2 ships as open weights on Hugging Face underneath MIT, with safetensors in F32, BF16, and FP8 F8_E4M3.
- The mannequin is a compact MoE with 229B complete parameters and ~10B lively per token, which the cardboard ties to decrease reminiscence use and steadier tail latency in plan, act, confirm loops typical of brokers.
- Outputs wrap inside reasoning in
... - Reported outcomes cowl Terminal-Bench, (Multi-)SWE-Bench, BrowseComp, and others, with scaffold notes for reproducibility, and day-0 serving is documented for SGLang and vLLM with concrete deploy guides.
Editorial Notes
MiniMax M2 lands with open weights underneath MIT, a combination of specialists design with 229B complete parameters and about 10B activated per token, which targets agent loops and coding duties with decrease reminiscence and steadier latency. It ships on Hugging Face in safetensors with FP32, BF16, and FP8 codecs, and supplies deployment notes plus a chat template. The API paperwork Anthropic appropriate endpoints and lists pricing with a restricted free window for analysis. vLLM and SGLang recipes can be found for native serving and benchmarking. Total, MiniMax M2 is a really strong open launch.
Take a look at the API Doc, Weights and Repo. Be happy to take a look at our GitHub Page for Tutorials, Codes and Notebooks. Additionally, be happy to comply with us on Twitter and don’t overlook to hitch our 100k+ ML SubReddit and Subscribe to our Newsletter. Wait! are you on telegram? now you can join us on telegram as well.
Michal Sutter is a knowledge science skilled with a Grasp of Science in Information Science from the College of Padova. With a strong basis in statistical evaluation, machine studying, and information engineering, Michal excels at remodeling complicated datasets into actionable insights.
