🔍 Read the full analysis: MiniMax M3.1 Flash: Our Newsroom Results Put Its Leaderboard Rank In Context 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 review reports a 43 score on Artificial Analysis’s Intelligence Index and describes a newsroom test of 24 articles, but the supplied material cuts off before the test results, so its performance there cannot be reported.
MiniMax released M3.1 Flash on October 19, pricing the model at $0.15 per million input tokens and $0.60 per million output tokens. A review published by ThorstenMeyerAI.com on October 20 compared MiniMax’s claims with Artificial Analysis data and a 24-article newsroom evaluation; the supplied review material does not include the newsroom scores, leaving that part of the comparison unresolved.
MiniMax offers M3.1 Flash through its API, M Plan subscription, MiniMax Code and OpenRouter. OpenRouter listed the model free for its first 72 hours. The listed price for cached input is $0.015 per million tokens. MiniMax says the new model costs half as much per token as M3 and about 70% less per agent task, though the review provides no independent task-cost comparison to verify that claim.
The company describes M3.1 Flash as a fast, low-cost model for agent work and says it can exceed 200 tokens per second in everyday use, with a median above 170. Artificial Analysis measured an output speed of 120 tokens per second and 1.08 seconds to first token in its evaluation. Those are different measurements and conditions; the source does not establish that one directly disproves the other.
The review says MiniMax’s published charts place M3.1 Flash ahead on Automation-Bench, while DeepSeek V4.1 Flash 1017 leads on DeepSWE and Terminal-Bench among the listed systems. Artificial Analysis gave M3.1 Flash an Intelligence Index score of 43, below several open-weight models in its comparison. The review also says the model used 180 million tokens to complete the index, compared with a 99-million-token median.
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 Meets Measured Performance
The release puts a lower-priced model into a crowded field where buyers must weigh more than the charge per token. Output speed, task quality and token use can all affect the practical cost of running an agent. Artificial Analysis’s reported token usage suggests that a cheaper rate alone may not translate into a cheaper completed task, particularly if the model consumes more tokens to finish an evaluation.
The gap between vendor claims, benchmark charts and independent measurements is relevant to developers deciding whether to deploy a model. MiniMax’s results are company-reported, according to the review, while Artificial Analysis evaluated the model using its own index. Neither source, on the supplied information, gives a complete account of performance across every production workload. The newsroom test could have offered a further practical comparison, but its scores are absent from the material provided.
As an affiliate, we earn on qualifying purchases.
From Space Bunny to M3.1 Flash
Before the official release, the model appeared anonymously on OpenRouter as Space Bunny Alpha. The review says it went live on September 23, led OpenRouter’s usage ranking for 10 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; that figure is the company’s account, as reported by the review.
MiniMax has not yet released the model weights. It said they would arrive on October 23, and the review reports that no license or official parameter count had been published as of October 20. Artificial Analysis consequently listed M3.1 Flash as proprietary at that time. The review also notes that MiniMax did not include DeepSeek in its chart for Agents’ Last Exam, despite making a ranking claim for Chinese models; DeepSeek’s model card reports a different score, possibly using a different setup.
““nearly three times faster” than M3”
— MiniMax, as quoted in the ThorstenMeyerAI.com review
low-cost AI chatbot development tools
As an affiliate, we earn on qualifying purchases.
As an affiliate, we earn on qualifying purchases.
Newsroom Scores Are Missing
The supplied review material describes a newsroom test using 12 news sources in English and German, producing 24 articles, with GPT-6.1 Sol and Claude Opus 5.5 as judges. It says the judges assessed source fidelity, invented facts, clarity, usefulness, structure and polish. But the source cuts off before presenting any scores or model-by-model results. It is not possible to state how M3.1 Flash performed in that test or compare it with other models based on the material available here.
Other points remain open. MiniMax’s speed and agent-cost statements are company claims; its benchmark charts are self-reported, and the review says they have not been independently reproduced. MiniMax has not confirmed the parameter count reported by another outlet, and the review says it has not explained what happened to prompts retained during the anonymous Space Bunny period. The model’s eventual license and the conditions of the promised weight release were also not available as of October 20.
AI model speed optimization hardware
As an affiliate, we earn on qualifying purchases.
As an affiliate, we earn on qualifying purchases.
Weights and Test Results Ahead
The next stated milestone is the planned October 23 weight release. Developers and reviewers will be able to check whether MiniMax meets that date and examine the license and model details when they are published. Further independent testing could clarify how its speed, token use and task performance compare with alternatives under consistent conditions.
The newsroom comparison should not be treated as evidence of a result until its missing scores and method details are available. For now, readers can assess the stated price, MiniMax’s claims and the Artificial Analysis measurements, while keeping the review’s practical news-writing results unresolved.
artificial intelligence benchmark tools
As an affiliate, we earn on qualifying purchases.
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 ThorstenMeyerAI.com review published the following day.
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. MiniMax’s claim that it costs about 70% less per agent task is not independently verified in the supplied material.
How did M3.1 Flash score on Artificial Analysis?
Artificial Analysis gave it an Intelligence Index score of 43. The review lists several models with higher scores and reports a measured output speed of 120 tokens per second.
What did the newsroom test find?
The review describes a test of 24 articles based on 12 sources in English and German, judged by GPT-6.1 Sol and Claude Opus 5.5. The supplied source material does not include the scores, so the test’s findings cannot be stated.
Are the model weights available?
Not as of the review’s October 20 publication date. MiniMax said the weights were expected on October 23; their release, license and final details were not confirmed in the supplied material.
Source: ThorstenMeyerAI.com
Fall Picks
fall essentials
As an affiliate, we earn on qualifying purchases.
