MiniMax M2.7 is your model for fixing code, reading error logs and building reports, budgets and slides. It scored 89.8 on GPQA Diamond and 94.2 on AIME 2026, so it handles hard questions too. MiniMax M2.7 is available for free on AIFreeForever.
MiniMax M2.7 is MiniMax's open coding and office model, released on 18 March 2026. It fixes code, reads error logs and edits Word, Excel and PowerPoint files over several rounds. Chat with MiniMax M2.7 free on AIFreeForever.
AIFreeForever is not affiliated with or endorsed by the model providers. All trademarks and logos belong to their owners.
Scores 56.22% on SWE-Pro, a test of fixing real code.
Scores 1495 on GDPval-AA, a test of office work.
Scores 89.8 on GPQA Diamond, a test of graduate-level science.
Scores 94.2 on MathArena AIME 2026, a maths contest test.
and choose MiniMax M2.7.
MiniMax M2.7 reads text only.
and ask for changes.
Plain chat
Plain chat
@Word
@Excel
@PPT
@Website AIFreeForever's free chatbot is fully integrated with MiniMax M2.7 via the official API providers. To get started, type your question or paste a supporting document, and MiniMax M2.7 answers in the same chat. Use it to fix a bug, find the cause of an error log, write a report or budget and more.
Our chatbot gives you MiniMax M2.7 free, without a MiniMax account or Token Plan.
It scored 56.22% on SWE-Pro, a test of fixing real code.
Our chatbot reads your error log and finds the likely cause.
Our chatbot turns answers into Word, Excel and PowerPoint files you can edit.
It scored 1495 on GDPval-AA, a test of office work.
MiniMax built it to keep a persona steady over a long chat.
MiniMax M2.7 works by planning a task and calling tools one after another. MiniMax used an early version of the model to improve its own coding setup over more than 100 rounds, as described in MiniMax's M2.7 model card.
MiniMax M2.7 scores highest on contest maths (94.2 on MathArena AIME 2026) and graduate-level science (89.8 on GPQA Diamond).
This benchmark was extracted from MiniMax.
On MLE Bench Lite, a test of machine-learning contests, MiniMax M2.7 got a 66.6% medal rate. Claude Opus 4.6 got 75.7%, GPT-5.4 got 71.2% and Gemini 3.1 got 66.6%, in MiniMax's M2.7 announcement.
MiniMax M2.7 is the model below built for editing office files, and it reads text only.
Pick MiniMax M2.7 for code fixes and office files. Pick GLM 5.3 for very long code and security checks.
MiniMax M2.7 costs $0.21 per million input tokens and $0.84 per million output tokens on OpenRouter. It's free on AIFreeForever.
Yes. You can use MiniMax M2.7 free in two ways:
MiniMax M2.7 has four limits on AIFreeForever, which connects to it through its API and adds our skills and file editors:
These limitations only cover what you can do with MiniMax M2.7 on AIFreeForever, not the limits of the model on MiniMax’s own platform. All trademarks belong to their owners, and the model’s full features are only available when you use it officially through MiniMax.
Yes, on AIFreeForever. On OpenRouter it costs $0.21 per million input tokens and $0.84 per million output tokens.
MiniMax made MiniMax M2.7.
MiniMax released M2.7 on 18 March 2026.
No. MiniMax M2.7 reads and writes text only.
It's open weights under a non-commercial licence. Personal and research use is free; business use needs MiniMax's written permission.
MiniMax M2.7 has 229B parameters.
No. It can make mistakes, so test its code and check important facts.
Ask MiniMax M2.7 a question, paste your code or add a file and get instant answers. Not satisfied? Compare your answers to other models for free.
Chat with MiniMax M2.7Tell us if this AI content is offensive, harmful or wrong. We review every report and use it to improve our filters.