The first entry is always the pin's source. Overall confidence is a weighted average of how firmly each reference supports the start and end times used above; a reference counts half as much for every 180 days older than the newest.
MiniMax's news post is dated 2025.10.27 and announces M2 as open-sourced with its API and MiniMax Agent free for a limited time; VentureBeat[3] covered it at 12:01 pm PT on 27 October, while the Hugging Face[5] weights were uploaded on 25 October UTC and Artificial Analysis[4] gives 26 October.
Technical paper for the M2 series (submitted 26 May 2026): the flagship M2 has 229.9B total and 9.8B activated parameters, trained with agent-driven data pipelines and the Forge reinforcement-learning system.[1]
VentureBeat (Carl Franzen), 27 Oct 2025: MIT-licensed sparse MoE with 230B total and 10B active parameters, first among open-weight models on the Artificial Analysis[4] Intelligence Index v3.0 with 61 points, servable on four H100s at FP8, $0.30 and $1.20 per million tokens, OpenAI and Anthropic API compatible.
Artificial Analysis: open-weights text model, 230B total and 10B active, released October 2025 (26 October per its FAQ), modified MIT license, $0.30 and $1.20 per million tokens, 205k context; now marked deprecated in favour of M2.1.
Model card: 230B total and 10B active parameters, coding and agentic tables (SWE-bench Verified 69.4, Terminal-Bench 46.3, BrowseComp 44, tau2-Bench 77.2), Artificial Analysis[4] intelligence table with AA Intelligence 61, interleaved-thinking requirement, SGLang, vLLM, MLX and Transformers deployment.
Something missing or wrong? Say it in your own words: a link that backs this pin up, a different start or end date and why, or a fact it lacks or gets wrong. The AI checks it against this pin's sources, searches for better ones, and adds any page that backs you up. The pin's own sources still count most. A picture that shows something else, or shows it badly, is looked at too, and moved down or replaced.
MiniMax's coding and agentic table from the model card; "-" means not reported, "*" marks figures taken from the rival's own report[5].
| Benchmark | MiniMax-M2 | Claude Sonnet 4 | Claude Sonnet 4.5 | Gemini 2.5 Pro | GPT-5 (thinking) | GLM-4.6 | Kimi K2 0905 | DeepSeek-V3.2 |
|---|---|---|---|---|---|---|---|---|
| SWE-bench Verified | 69.4 | 72.7* | 77.2* | 63.8* | 74.9* | 68* | 69.2* | 67.8* |
| Multi-SWE-Bench | 36.2 | 35.7* | 44.3 | - | - | 30 | 33.5 | 30.6 |
| SWE-bench Multilingual | 56.5 | 56.9* | 68 | - | - | 53.8 | 55.9* | 57.9* |
| Terminal-Bench | 46.3 | 36.4* | 50* | 25.3* | 43.8* | 40.5* | 44.5* | 37.7* |
| ArtifactsBench | 66.8 | 57.3* | 61.5 | 57.7* | 73* | 59.8 | 54.2 | 55.8 |
| BrowseComp | 44 | 12.2 | 19.6 | 9.9 | 54.9* | 45.1* | 14.1 | 40.1* |
| BrowseComp-zh | 48.5 | 29.1 | 40.8 | 32.2 | 65 | 49.5 | 28.8 | 47.9* |
| GAIA (text only) | 75.7 | 68.3 | 71.2 | 60.2 | 76.4 | 71.9 | 60.2 | 63.5 |
| xbench-DeepSearch | 72 | 64.6 | 66 | 56 | 77.8 | 70 | 61 | 71 |
| HLE (w/ tools) | 31.8 | 20.3 | 24.5 | 28.4* | 35.2* | 30.4* | 26.9* | 27.2* |
| tau2-Bench | 77.2 | 65.5* | 84.7* | 59.2 | 80.1* | 75.9* | 70.3 | 66.7 |
| FinSearchComp-global | 65.5 | 42 | 60.8 | 42.6* | 63.9* | 29.2 | 29.5* | 26.2 |
| AgentCompany | 36 | 37 | 41 | 39.3* | - | 35 | 30 | 34 |