MiniMax model product
MiniMax M2 is an open-source large language model by MiniMax, built for agents and coding tasks. It delivers state-of-the-art tool use, reasoning, and search performance while maintaining exceptional cost-efficiency and speed, priced at just 8% of Claude 3.5 Sonnet’s cost and running at nearly double its inference speed (≈100 TPS). Designed for end-to-end agentic workflows, it excels at long-chain tool calling across Shell, Browser, Python, and other MCP tools. While slightly behind top overseas models in programming, it ranks among the best domestic models and top five globally on the Artificial Analysis benchmark. M2 powers the MiniMax Agent platform, available in Lightning Mode for fast tasks and Pro Mode for complex multi-step reasoning, and its weights, API, and deployment guides are freely available on Hugging Face, vLLM, and SGLang.
Updated Aug 10, 2026. Default version: MiniMax M2
Structured fields from the published default version.
This profile uses the latest version under this unique model that has a calculated LLMBoard score. Arena and price are excluded.
Published benchmark records for the scored version MiniMax M2.
| IF | 0.7 | 2 | 2 | 0.0% | C | |
| AA-Index | 0.6 | 3 | 3 | 0.0% | C | |
| Terminal-Bench | 0.5 | 4 | 25 | 87.5% | C | |
| LiveCodeBench | 0.8 | 5 | 73 | 94.4% | C | |
| Multi-SWE-Bench | 0.4 | 5 | 6 | 20.0% | C | |
| Tau-bench | 0.8 | 5 | 6 | 20.0% | C | |
| BrowseComp-zh | 0.5 | 11 | 13 | 16.7% | C | |
| SciCode | 0.4 | 17 | 18 | 5.9% | C | |
| Tau2 Telecom | 0.9 | 17 | 35 | 52.9% | C | |
| SWE-bench Multilingual | 0.6 | 27 | 34 | 21.2% | C | |
| MMLU-Pro | 0.8 | 39 | 129 | 70.3% | C | |
| BrowseComp | 0.4 | 50 | 58 | 14.0% | C | |
| SWE-Bench Verified | 0.7 | 62 | 104 | 40.8% | C | |
| AIME 2025 | 0.8 | 75 | 114 | 34.5% | C | |
| Humanity's Last Exam | 0.1 | 76 | 92 | 17.6% | C | |
| GPQA | 0.8 | 94 | 233 | 59.9% | C |
Preference and agent-evaluation signals from published Arena datasets.
| webdev | webdev-html | 89 | 1306.8 | 6,473 | N/A | |
| webdev | overall | 90 | 1296.7 | 6,502 | N/A | |
| webdev | webdev | 90 | 1296.7 | 6,502 | N/A | |
| text factuality | industry business and management and financial operations | 125 | 1375.6 | 758 | N/A | |
| text factuality | industry entertainment and sports and media | 125 | 1322.6 | 602 | N/A | |
| text factuality | coding | 130 | 1418.8 | 873 | N/A | |
| text factuality | industry writing and literature and language | 131 | 1317.4 | 846 | N/A | |
| text factuality | hard prompts english | 133 | 1402.9 | 1,064 | N/A | |
| text factuality | instruction following | 133 | 1346.2 | 1,126 | N/A | |
| text factuality | longer query | 134 | 1359.1 | 900 | N/A | |
| text factuality | industry software and it services | 135 | 1406.8 | 1,433 | N/A | |
| text factuality | hard prompts | 141 | 1389.2 | 3,399 | N/A | |
| text factuality | english | 144 | 1387.4 | 2,938 | N/A | |
| text factuality | exclude ties | 146 | 1334.7 | 4,819 | N/A | |
| text factuality | non english | 147 | 1335.2 | 3,137 | N/A | |
| text factuality | overall | 149 | 1362.8 | 6,692 | N/A | |
| text | multi turn | 157 | 1367.2 | 1,116 | N/A | |
| text | chinese | 162 | 1384.3 | 278 | N/A | |
| text | industry mathematical | 164 | 1367.2 | 286 | N/A | |
| text | industry medicine and healthcare | 165 | 1365.8 | 319 | N/A | |
| text | hard prompts english | 169 | 1375.2 | 1,902 | N/A | |
| text | polish | 169 | 1316.1 | 266 | N/A | |
| text | coding | 170 | 1373.7 | 1,545 | N/A | |
| text | industry software and it services | 170 | 1376.5 | 2,468 | N/A | |
| text style control | industry mathematical | 170 | 1369.6 | 286 | N/A | |
| text | spanish | 173 | 1319.0 | 173 | N/A | |
| text style control | chinese | 173 | 1386.9 | 278 | N/A | |
| text style control | multi turn | 173 | 1363.9 | 1,116 | N/A | |
| text | hard prompts | 174 | 1359.1 | 3,691 | N/A | |
| text | instruction following | 174 | 1333.7 | 1,983 | N/A | |
| text | industry business and management and financial operations | 177 | 1338.0 | 1,314 | N/A | |
| text | russian | 177 | 1328.3 | 323 | N/A | |
| text | industry legal and government | 181 | 1346.1 | 429 | N/A | |
| text | longer query | 181 | 1335.6 | 1,582 | N/A | |
| text | english | 182 | 1364.2 | 3,363 | N/A | |
| text | overall | 183 | 1342.4 | 6,859 | N/A | |
| text | exclude ties | 183 | 1299.3 | 4,998 | N/A | |
| text style control | hard prompts english | 183 | 1387.7 | 1,902 | N/A | |
| text style control | math | 185 | 1353.7 | 320 | N/A | |
| text style control | polish | 185 | 1310.1 | 266 | N/A | |
| text | expert | 186 | 1320.9 | 256 | N/A | |
| text | industry life and physical and social science | 187 | 1349.3 | 1,108 | N/A | |
| text | math | 187 | 1351.0 | 320 | N/A | |
| text style control | hard prompts | 187 | 1368.7 | 3,691 | N/A | |
| text | industry entertainment and sports and media | 188 | 1302.7 | 1,175 | N/A | |
| text style control | english | 189 | 1370.8 | 3,363 | N/A | |
| text style control | industry software and it services | 189 | 1386.0 | 2,468 | N/A | |
| text style control | industry business and management and financial operations | 190 | 1349.5 | 1,314 | N/A | |
| text | non english | 191 | 1315.8 | 3,448 | N/A | |
| text style control | industry medicine and healthcare | 191 | 1364.0 | 319 | N/A | |
| text style control | instruction following | 191 | 1338.9 | 1,983 | N/A | |
| text style control | russian | 192 | 1332.9 | 323 | N/A | |
| text style control | spanish | 193 | 1310.6 | 173 | N/A | |
| text style control | industry legal and government | 195 | 1357.0 | 429 | N/A | |
| text style control | exclude ties | 199 | 1303.6 | 4,998 | N/A | |
| text style control | expert | 199 | 1337.9 | 256 | N/A | |
| text style control | overall | 200 | 1346.0 | 6,859 | N/A | |
| text style control | longer query | 201 | 1343.2 | 1,582 | N/A | |
| text | creative writing | 202 | 1290.6 | 892 | N/A | |
| text style control | coding | 202 | 1384.7 | 1,545 | N/A | |
| text | industry writing and literature and language | 203 | 1299.1 | 1,467 | N/A | |
| text style control | industry entertainment and sports and media | 203 | 1302.5 | 1,175 | N/A | |
| text style control | industry life and physical and social science | 203 | 1356.6 | 1,108 | N/A | |
| text style control | non english | 206 | 1316.3 | 3,448 | N/A | |
| text style control | creative writing | 226 | 1288.5 | 892 | N/A | |
| text style control | industry writing and literature and language | 230 | 1302.1 | 1,467 | N/A |
Official vendor API PAYG pricing is summarized first. The table then lists individual provider offerings without treating their minimum as the official price.
| MiniMax Token Plan (minimax.io) | MiniMax-M2 | global | N/A | N/A | 196.6K | |
| MiniMax Token Plan (minimaxi.com) | MiniMax-M2 | global | N/A | N/A | 196.6K | |
| NanoGPT | MiniMax-M2 | global | $0.17 | $1.53 | 200K | |
| LLM Gateway | minimax-m2 | global | $0.20 | $1 | 196.6K | |
| MiniMax (minimax.io) | MiniMax-M2 | global | $0.30 | $1.2 | 196.6K | |
| MiniMax (minimaxi.com) | MiniMax-M2 | global | $0.30 | $1.2 | 196.6K | |
| 302.AI | MiniMax-M2 | global | $0.33 | $1.32 | 1M | |
| Cortecs | minimax-m2 | global | $0.349 | $1.41 | 400K |
Official prices use only the vendor's configured official Provider and positive standard USD PAYG rates. Third-party offers remain explicitly labeled.
All published versions linked to this unique model. The score columns identify the version used by the current overall ranking.
| MiniMax M2 | 44.9 | 230B | 196.6K | 128K | Yes | MIT |
A concise description based on the published model registry.
MiniMax M2 is an open-source large language model by MiniMax, built for agents and coding tasks. It delivers state-of-the-art tool use, reasoning, and search performance while maintaining exceptional cost-efficiency and speed, priced at just 8% of Claude 3.5 Sonnet’s cost and running at nearly double its inference speed (≈100 TPS). Designed for end-to-end agentic workflows, it excels at long-chain tool calling across Shell, Browser, Python, and other MCP tools. While slightly behind top overseas models in programming, it ranks among the best domestic models and top five globally on the Artificial Analysis benchmark. M2 powers the MiniMax Agent platform, available in Lightning Mode for fast tasks and Pro Mode for complex multi-step reasoning, and its weights, API, and deployment guides are freely available on Hugging Face, vLLM, and SGLang.
Use the benchmark, Arena and pricing sections above as separate evidence. A missing field means the current data snapshot does not support that claim.
Data snapshot: 2026-08-07. Editorial model content is not available in the backend.
Open a comparison with the three ranked models immediately above and below this model.
Recommendations prioritize the same model type and family, then the closest published LLMBoard score.
Common questions about MiniMax M2.
MiniMax M2's default version was released on Oct 27, 2025.
MiniMax M2's official API price is $0.30 per million input tokens and $1.2 per million output tokens via MiniMax (minimax.io). The lowest tracked third-party offer starts at $0.17 input and $1.53 output via NanoGPT.
MiniMax M2 is published under MiniMax in the model registry.
The default version has a 196.6K token context window.
Yes. The default version is marked as open weight under MIT.
8 published provider offerings are linked to the default version.
Nearby ranked alternatives include Gemma 3 27B, Nemotron 3 Nano, Granite 3.3 8B.