🔍 Read the full analysis: What Our Newsroom Test Revealed About MiniMax M3.1 Flash on ThorstenMeyerAI.com
Get business pricing on monitors, keyboards and dev gear
- Business-only prices and quantity discounts
- Tax-exempt purchasing
- Multiple users, one account, clear invoices
TL;DR
MiniMax released M3.1 Flash on October 19, 2026, pricing it at $0.15 per million input tokens and $0.60 per million output tokens. The source report compares MiniMax’s claims with Artificial Analysis measurements and a newsroom test of 24 articles, but it provides no newsroom scores or results table, so that test’s outcome cannot be reported here.
MiniMax says M3.1 Flash is designed for **fast, lower-cost agent work** and costs half as much per token as its predecessor, M3. The company also claims the model is nearly three times faster than M3 in everyday use, with speeds above 200 tokens per second and a median above 170. Those figures are **MiniMax’s claims**, not independent measurements.
Artificial Analysis, which evaluates models using a common set of ten tests, gave M3.1 Flash an **Intelligence Index score of 43**. That was below listed open-weight competitors including MiMo-V2.6-Pro at 46 and DeepSeek V4.1 Flash 1017 and GLM-5.3 at 45. Artificial Analysis measured output speed at **120 tokens per second** and time to first token at 1.08 seconds, below MiniMax’s speed claim. It also recorded 180 million tokens used to complete its index, compared with a 99 million-token median.
The newsroom review says it tested 12 real news sources in English and German, producing **24 articles**. Two judges, GPT-6.1 Sol and Claude Opus 5.5, scored each article against its source on fidelity, fabricated details, clarity, usefulness, structure and polish. The article says the model ran through OpenRouter’s free endpoint with a 32,000-token output limit, but the supplied material ends before reporting scores or comparisons. Its newsroom findings therefore cannot be stated.
Infographic 01 · MiniMax M3.1 Flash
A step behind on the index. Ahead of every open model in our newsroom.
An independent intelligence index puts MiniMax M3.1 Flash a step behind the open-weight leaders. Writing real news from real sources, it beat all of them.
Independent index
Artificial Analysis Intelligence Index, 10 evaluations. Bar length on a 0–50 scale.
Artificial Analysis lists M3.1 Flash as proprietary until its weights are released.
Our newsroom test
Mean score, English articles, 1–5 by GPT-6.1 Sol and Claude Opus 5.5 with the source in context. Bar length on a 0–5 scale.
In German, M3.1 Flash (3.36) also leads every open model, but GPT-6 Luna is clearly ahead (3.62).
Sources: Artificial Analysis, MiniMax-M3.1-Flash and open-weights models (20 Oct 2026); Thorsten Meyer AI newsroom benchmark, 12 sources × English + German, run 20 Oct 2026.
Powered by Thorsten Meyer AI
Price and Speed Against Quality
M3.1 Flash’s **low listed token price** could appeal to developers running agents or processing large volumes of text. But price alone does not establish the cost of a completed task: Artificial Analysis found the model used **180 million tokens** in its index run, against a 99 million-token median. More tokens can offset a lower per-token rate, though the supplied data does not calculate total task costs.
The gap between MiniMax’s speed claim and the independent measurement also matters for buyers weighing latency. Artificial Analysis measured **120 tokens per second**, while MiniMax cited more than 200 in everyday use. Different test conditions may affect results, and the report does not supply enough methodological detail to reconcile them. The newsroom benchmark could offer a practical measure of source fidelity, but its missing results leave that question open.
AI language model performance tester
As an affiliate, we earn on qualifying purchases.
As an affiliate, we earn on qualifying purchases.
From Space Bunny to M3.1 Flash
Before its public release, the model appeared anonymously on OpenRouter as **Space Bunny Alpha** beginning September 23. ThorstenMeyerAI.com reports that it led the platform’s usage ranking for ten days and processed **56.7 trillion tokens** there before being removed on October 5. MiniMax says the model processed more than 200 trillion tokens overall.
The report says MiniMax revealed the model’s identity on October 19. It is available through the MiniMax API, M Plan subscription, MiniMax Code and OpenRouter; OpenRouter offered it free for its first 72 hours. Its listed context window is **one million tokens**, while OpenRouter allows up to 524,288 output tokens. The report says the model’s weights had not yet been released and MiniMax promised them for October 23.
MiniMax’s published comparison charts show M3.1 Flash ahead of listed rivals on Automation-Bench, but DeepSeek V4.1 Flash 1017 leads on DeepSWE and Terminal-Bench. The source report cautions that these are **company-reported benchmark results** and that MiniMax selected the competitors. It also says some comparisons omit newer models and that MiniMax’s internal security results have not been independently reproduced.
“More than 200 tokens per second in everyday use, with a median above 170.”
— MiniMax
As an affiliate, we earn on qualifying purchases.
Newsroom Scores Are Missing
The source material does not include the newsroom benchmark’s **scores, rankings or comparison table**. It is not possible to say whether M3.1 Flash produced accurate articles, how it performed in English versus German, or how it compared with other models. The test design is described, but its results are not.
Other open questions include whether the weights will be released on the promised date, what license and parameter count will accompany them, and whether independent evaluators can reproduce MiniMax’s benchmark and security claims. The source says MiniMax has not confirmed a parameter count; a figure reported elsewhere is explicitly unconfirmed. It also says MiniMax has not clarified what happened to prompts retained during the anonymous OpenRouter period, when the provider could keep prompts but was not permitted to train on them.
As an affiliate, we earn on qualifying purchases.
Weights and Test Results to Watch
The next stated milestone is **October 23**, when MiniMax says it will release M3.1 Flash’s weights. Their publication would allow observers to check the model’s license and disclosed specifications, and could change its status from proprietary in Artificial Analysis’s listing.
The newsroom review’s missing results are also needed before its practical conclusion can be reported. Further independent testing could clarify whether the model’s low token price translates into lower costs for completed tasks, and how its output quality compares with alternatives. Until then, the confirmed picture is a competitively priced release with an external index score of 43, an independently measured speed below MiniMax’s claim, and an unpublished newsroom-test outcome.
As an affiliate, we earn on qualifying purchases.
Key Questions
When did MiniMax release M3.1 Flash?
MiniMax released **M3.1 Flash on October 19, 2026**, according to the source report.
How much does M3.1 Flash cost?
The listed price is **$0.15 per million input tokens** and **$0.60 per million output tokens**. Cached input is listed at $0.015 per million tokens.
How fast is the model?
MiniMax claims more than 200 tokens per second in everyday use and a median above 170. Artificial Analysis measured **120 tokens per second**. The source does not establish whether the tests used the same conditions.
What did the newsroom test find?
The supplied source describes a test of **24 articles** in English and German, judged by GPT-6.1 Sol and Claude Opus 5.5. It does not include the scores or findings, so the test’s outcome cannot be confirmed from the material provided.
Are M3.1 Flash’s weights available?
They had not been released when the report was published on October 20. MiniMax said the weights would arrive on **October 23, 2026**; the source does not confirm whether that release occurred.
Source: ThorstenMeyerAI.com
Halloween Picks
halloween
As an affiliate, we earn on qualifying purchases.
