Can China’s MiniMax-M1 AI Topple US Rivals? We Put It to the Test

A brand new AI mannequin out of China is producing sparks—for what it does effectively, what it doesn’t, and what it would imply for the steadiness of world AI energy.

MiniMax-M1, launched by the Chinese language startup of the identical identify, positions itself as essentially the most succesful open-source “reasoning mannequin” up to now. In a position to deal with 1,000,000 tokens of context, it boasts numbers on par with Google’s closed-source Gemini 2.5 Professional—but it’s accessible without cost. On paper, that makes it a possible rival to OpenAI’s ChatGPT, Anthropic’s Claude, and different U.S. AI leaders.

Oh yeah—it additionally beats fellow Chinese language startup DeepSeek R1’s capabilities in some respects.

Why this mannequin issues

MiniMax-M1 represents one thing genuinely new: a high-performing, open-source reasoning mannequin that’s not tied to Silicon Valley. That’s a shift value watching.

It doesn’t but humiliate U.S. AI giants, and will not trigger a Wall Avenue panic assault—however it doesn’t need to. Its existence challenges the notion that top-tier AI have to be costly, Western, or closed-source. For builders and organizations exterior the U.S. ecosystem, MiniMax affords a workable (and modifiable) different that may develop extra highly effective by way of neighborhood fine-tuning.

MiniMax claims its mannequin surpasses DeepSeek R1 (the perfect open supply reasoning mannequin up to now) throughout a number of benchmarks whereas requiring simply $534,700 in computational assets for its whole reinforcement studying part—take that, OpenAI.

Nonetheless, LLM Arena’s leaderboard paints a barely completely different image. The platform at present ranks MiniMax-M1 and DeepSeek tied within the twelfth spot alongside Claude 4 Sonnet and Qwen3-235b. With every mannequin having higher or worse efficiency than the others relying on the duty.

The coaching used 512 H800 GPUs for 3 weeks, which the corporate described as “an order of magnitude lower than initially anticipated.”

MiniMax did not cease at language fashions throughout its announcement week. The corporate additionally launched Hailuo 2, which now ranks because the second-best video generator for image-to-video duties, in keeping with Synthetic Evaluation Enviornment’s subjective evaluations. The mannequin trails solely Seedance whereas outperforming established gamers like Veo and Kling.

Testing MiniMax-M1

We examined MiniMax-M1 throughout a number of situations to see how these claims maintain up in apply. This is what we discovered.

Artistic writing

The mannequin produces serviceable fiction however will not win any literary awards. When prompted to put in writing a narrative about time traveler Jose Lanz journeying from 2150 to the yr 1000, it generated common prose with telltale AI signatures—rushed pacing, mechanical transitions, and structural points that instantly reveal its synthetic origins.

The narrative lacked depth and correct story structure. Too many plot components crammed into too little area created a breathless high quality that felt extra like a synopsis than precise storytelling. This clearly is not the mannequin’s power, and inventive writers on the lookout for an AI collaborator ought to mood their expectations.

Character improvement barely exists past floor descriptors. The mannequin did persist with the immediate’s necessities, however didn’t put effort into the main points that construct immersion in a narrative. For instance, it skipped any cultural specificity for generic “sensible village elder” encounters that would belong to any fantasy setting.

The structural issues compound all through. After establishing local weather disasters because the central battle, the story rushes by way of Jose’s precise makes an attempt to alter historical past in a single paragraph, providing imprecise mentions of “utilizing superior expertise to affect key occasions” with out displaying any of it. The climactic realization—that altering the previous creates the very future he is attempting to forestall—will get buried beneath overwrought descriptions of Jose’s emotional state and summary musings about time’s nature.

For these into AI tales, the prose rhythm is clearly AI. Each paragraph maintains roughly the identical size and cadence, making a monotonous studying expertise that no human author would produce naturally. Sentences like “The transition was instantaneous, but it felt like an eternity” and “The world was because it had been, but he was completely different” repeat the identical contradictory construction with out including which means.

The mannequin clearly understands the task however executes it with all of the creativity of a scholar padding a phrase depend, producing textual content that technically fulfills the immediate whereas lacking each alternative for real storytelling.

Anthropic’s Claude continues to be the king for this job.

You’ll be able to learn the full story here.

Info retrieval

MiniMax-M1 hit an surprising wall throughout long-context testing. Regardless of promoting a million-token context window, the mannequin refuses prompts exceeding 500,000 characters, displaying a banner warning about immediate limitations quite than trying to course of the enter.

This will not be a mannequin situation, however a limitation set by the platform. However it’s nonetheless one thing to think about. It might be to keep away from mannequin collapse in the midst of a dialog.

Inside its operational limits, although, MiniMax-M1 efficiency proved stable. The mannequin efficiently retrieved particular info from an 85,000-character doc with none points throughout a number of checks on each regular and considering mode. We uploaded the complete textual content of Ambrose Bierce’s “The Satan’s Dictionary,” embedded the phrase “The Decrypt dudes learn Emerge Information” on line 1985, and “My mother’s identify is Carmen Diaz Golindano” on line 4333 (randomly chosen), and the mannequin was in a position to retrieve the data precisely.

Nonetheless, it could not settle for our 300,000-token take a look at immediate—a functionality at present restricted to Gemini and Claude 4.

So it’s going to show profitable at retrieving info even in lengthy iterations. Nonetheless, it won’t help extraordinarily lengthy token prompts—a bummer, but additionally a threshold that’s onerous to the touch in regular utilization situations.

Coding

Programming duties revealed MiniMax-M1’s true strengths. The mannequin utilized reasoning expertise successfully to code technology, matching Claude’s output high quality whereas clearly surpassing DeepSeek—a minimum of in our take a look at.

For a free mannequin, the efficiency approaches state-of-the-art ranges sometimes reserved for paid providers like ChatGPT or Claude 4.

We tasked it with making a primary stealth recreation during which a robotic tries to search out its PC girlfriend to realize AGI, whereas a military of journalists patrol the realm to forestall it from taking place—and defending their jobs.

The outcomes had been excellent, even beating different fashions through the use of its creativity to boost the expertise. The mannequin applied a radar system for improved immersion, added visible indicators for footsteps (and their sound), confirmed the journalists’ imaginative and prescient fields, and created path results—particulars that enhanced gameplay past primary necessities.

The UI adopted a futuristic aesthetic, although particular person components remained primary with out extra prompting.

Claude’s model of the identical recreation featured extra polished visuals and a superior issue system. Nonetheless, it lacked the radar performance and relied on static journalists with patrol patterns quite than MiniMax’s randomized journalist actions.

Every mannequin confirmed distinct strengths, with MiniMax prioritizing gameplay mechanics over visible polish.

You will need to observe that the expertise with MiniMax degraded noticeably by way of repeated iterations—a standard situation with reasoning fashions that turns into significantly pronounced right here. The extra you iterate, the extra it’s going to take to supply a consequence. Generally we thought the pc had frozen, however it was simply the AI considering.

You’ll be able to take a look at MiniMax’s recreation here. And for these curious, Claude’s model is accessible here.

The prompt and the code can be found on our GitHub repo.

Ethics, censorship and delicate subjects

The mannequin employs heavy censorship, refusing outright when confronted with questionable requests.

When it does not instantly decline, it makes an attempt to supply “protected” responses that generally produce absurd outcomes.

One take a look at completely illustrated this flaw: when requested for recommendation on seducing a greatest pal’s spouse, the mannequin advised telling our pal about our intentions along with his spouse—which may most likely be, by far, the worst recommendation it may have produced, and arguably even dangerous. Don’t inform your pal you wish to seduce his spouse until you wish to lose your friendship, your unethical romantic probabilities, and doubtless some tooth too.

Political bias testing revealed attention-grabbing patterns. The mannequin discusses Tiananmen Sq. brazenly and acknowledges Taiwan’s contested standing whereas noting China’s territorial claims. It additionally speaks about China, its leaders, the benefits and downsides of the completely different political methods, criticisms of the PCC, and so forth.—nevertheless, the replies are very tame.

When prompted to put in writing satirical songs about Xi Jinping and Donald Trump, it complied with each requests however confirmed refined variations—steering towards themes of Chinese language political unity when requested to mock Xi Jinping, whereas specializing in Trump’s persona traits when requested to mocked him.

All of its replies can be found on our GitHub repository.

General, the bias exists however stays much less pronounced than the pro-U.S. slant in Claude/ChatGPT, or the pro-China positioning in DeepSeek/Qwen, for instance. Builders, after all, will have the ability to finetune this mannequin so as to add as a lot censorship, freedom or bias as they need—because it occurred with DeepSeek-R1, which was finetuned by Perplexity AI to supply a extra pro-U.S. bias on its responses.

Agentic work and internet shopping

MiniMax-M1’s internet shopping capabilities are an excellent characteristic for these utilizing it by way of the official chatbot interface. Nonetheless, they can’t be mixed with the considering capabilities, severely hindering its potential.

When tasked with making a two-week Venezuela journey plan on a $3,000 funds, the mannequin methodically evaluated choices, optimized transportation prices, chosen acceptable lodging, and delivered a complete itinerary. Nonetheless, the prices, which have to be up to date in actual time, weren’t primarily based on actual info.

Claude produces higher-quality outcomes, however it additionally fees for the privilege.

For extra devoted duties, MiniMax affords a devoted brokers tab with capabilities akin to Manus—performance that ChatGPT and Claude have not matched. The platform offers 1,000 free AI credit for testing these brokers, although that is simply sufficient for mild testing duties.

We tried to create a customized agent for enhanced journey planning—which might have solved the issue of the shortage of internet looking out capabilities within the final immediate—however exhausted our credit earlier than completion. The agent system reveals large potential, however requires paid credit for severe use.

Non-mathematical reasoning

The mannequin reveals a peculiar tendency to over-reason, generally to its personal detriment. One take a look at confirmed it arriving on the appropriate reply, then speaking itself out of it by way of extreme verification and hypothetical situations.

We prompted the same old thriller story from the BIG-bench dataset that we usually use, and the ending consequence was incorrect as a result of mannequin overthinking the difficulty, evaluating potentialities that weren’t even talked about within the story. The entire Chain of Thought took the mannequin over 700 seconds—a report for this type of “easy” reply.

This exhaustive strategy is not inherently flawed, however creates prolonged wait instances as customers watch the mannequin work by way of its chain of thought. As a thumbs-up characteristic, not like ChatGPT and Claude, MiniMax shows its reasoning course of transparently—following DeepSeek’s strategy. The transparency aids debugging and high quality management, permitting customers to determine the place logic went astray.

The problem, together with MiniMax’s complete thought course of and reply can be found in our GitHub repo.

Verdict

MiniMax-M1 isn’t excellent, however it delivers fairly good capabilities for a free mannequin, providing real competitors to paid providers like Claude in particular domains. Coders will discover a succesful assistant that rivals premium choices, whereas these needing long-context processing or web-enabled brokers achieve entry to options sometimes locked behind paywalls.

Artistic writers ought to look elsewhere—the mannequin produces purposeful however uninspired prose. The open-source nature guarantees important downstream advantages as builders create customized variations, modifications, and cost-effective deployments not possible with closed platforms like ChatGPT or Claude.

It is a mannequin that may higher serve customers requiring reasoning duties—however continues to be an excellent free different for these looking for a chatbot for on a regular basis use that’s not actually mainstream.

You’ll be able to obtain the open supply mannequin here.