Gemini 3.5 Flash vs MiniMax M3: Fast Proprietary vs Open-Weight Budget

Gemini 3.5 Flash and MiniMax M3 are the two large language models compared here, and they split the win cleanly. Gemini 3.5 Flash is Google DeepMind's generally available fast tier, launched at Google I/O on May 19, 2026, a closed and managed model priced at 1.50 dollars per million input tokens and 9.00 dollars per million output tokens across a 1,048,576-token context window. MiniMax M3, released June 1, 2026, is an open-weight mixture-of-experts model with 428 billion total parameters and about 23 billion active, natively multimodal, priced at 0.30 dollars input and 1.20 dollars output per million tokens for prompts up to 512K tokens, with the same 1,048,576-token window. On the one independent evaluator that scores both the same way, the Artificial Analysis Intelligence Index version 4.1, Gemini 3.5 Flash leads 50 to 44, and it is built for speed. MiniMax M3 is exactly five times cheaper on input and seven and a half times cheaper on output at standard rates, and it ships open weights you can self-host. This is a split verdict, not a single winner. Best for measured intelligence, raw speed, and flat context pricing: Gemini 3.5 Flash. Best for cost and open weights: MiniMax M3.

Quick Verdict

This is a split verdict by use case, not a single overall winner. We ran both models side by side through their APIs, took the pricing straight from each vendor, and added our own hands-on observations from using both on reasoning and coding prompts. We have not run weeks of controlled, identical-task benchmarking of the two against each other, so where we lean on numbers we attribute them to their source and keep independent scores strictly apart from vendor-reported ones. The honest summary is that these two are not really fighting for the same buyer: one is a fast, polished proprietary model, the other a cheap, open one. Here is the short version.

• Best for measured intelligence: Gemini 3.5 Flash. On the Artificial Analysis Intelligence Index version 4.1, the one composite that scores both models the same way, Gemini sits at 50 while MiniMax M3 scores 44, a clear six-point lead.
• Best for speed: Gemini 3.5 Flash. Speed is the whole point of the Flash tier, and Google built it to run several times faster than frontier-class models while holding frontier-adjacent quality.
• Best for cost: MiniMax M3, and it is not close at standard rates. Output at 1.20 dollars per million tokens is exactly seven and a half times cheaper than Gemini at 9.00 dollars, and input at 0.30 dollars is exactly five times cheaper than Gemini at 1.50 dollars.
• Best for open weights and self-hosting: MiniMax M3. The weights ship openly, so you can download, self-host, and fine-tune on your own hardware. Gemini 3.5 Flash is closed and available only as a managed Google service.
• Best for predictable long-context pricing: Gemini 3.5 Flash. Both models carry the same 1,048,576-token window, but Gemini charges one flat rate across it, while MiniMax doubles its rate above 512K tokens.

Bottom line: if you want the higher independent capability score, the fastest responses, or a flat rate that does not jump on long prompts, pick Gemini 3.5 Flash. If you are cost-constrained or want to own and self-host your weights, pick MiniMax M3. We did not crown a single winner because the two optimize for different things, and the numbers back both stories at once: a clear capability and speed edge for Gemini and a large price and openness edge for MiniMax.

At a Glance

Before the detail, here is the side-by-side that frames everything below. All pricing was taken directly from each vendor. The one independent capability figure comes from the Artificial Analysis Intelligence Index version 4.1, and any vendor-reported benchmark is kept out of this table and labeled separately in the text.

Overview of Each Model

Gemini 3.5 Flash

Gemini 3.5 Flash is Google DeepMind's generally available fast tier, launched at Google I/O on May 19, 2026. It is worth being precise about the name: this is not the earlier Gemini 3 Flash Preview, which is a cheaper and less capable model. Gemini 3.5 Flash is the full generally available release, positioned as a frontier-adjacent model that is engineered above all for speed. On the one independent evaluator that scores both models in this matchup the same way, the Artificial Analysis Intelligence Index version 4.1, it scores 50, the higher of the two here and a genuinely strong number for a fast, low-latency tier. Its defining trait is throughput: Google built the Flash line to run several times faster than frontier-class models, which makes it a natural fit for interactive products, high-volume pipelines, and anything latency-sensitive. It is natively multimodal, carries a 1,048,576-token context window, and prices that entire window at a single flat rate of 1.50 dollars per million input tokens and 9.00 dollars per million output tokens, with cached input far cheaper. Google also publishes its own benchmark figures for the model, and we label them as such: on Google-reported evaluations it lands around 76 percent on Terminal-Bench 2.1, around 84 percent on the MCP Atlas agentic benchmark, and around 84 percent on CharXiv. Those are vendor numbers, useful as directional signals rather than independently verified scores. The trade-offs are the ones inherent to a closed model: no downloadable weights, no self-hosting, and access only through Google. For the full breakdown, see our Gemini 3.5 Flash review, and for the launch context our report on Gemini 3.5 Flash at I/O 2026.

MiniMax M3

MiniMax M3 is MiniMax's open-weight flagship, released on June 1, 2026 by the Shanghai-based lab. It is a mixture-of-experts model with 428 billion total parameters and about 23 billion active per token, which is the engineering trick behind its aggressive pricing: activating only a slice of the network per token keeps inference cheap. It is natively multimodal, uses a sparse attention design MiniMax calls MSA to make its long context affordable to serve, and ships open weights, so you can download, self-host, fine-tune, and redistribute the model rather than renting it through an API. Standard pricing is 0.30 dollars per million input tokens and 1.20 dollars per million output tokens for prompts up to 512K tokens, with a 1,048,576-token context window; note that the rate doubles for any prompt above 512K tokens, a detail that matters for long-context work. On the independent Artificial Analysis Intelligence Index version 4.1 it scores 44, which is a strong result for a model you can carry off and run on your own hardware, and it sits level with the other leading open-weight models of its generation. Our full MiniMax M3 review covers the architecture and licensing in more depth, and our launch write-up on the MiniMax M3 open-weight release has the background.

Pricing Compared

This is where the two models diverge most, and it is the single most important thing to understand about the matchup. We took every number below directly from each vendor rather than from secondhand summaries.

Run the arithmetic on the standard tier and the gap is large. On output, the number that dominates real agentic spend, MiniMax M3 at 1.20 dollars per million tokens is exactly seven and a half times cheaper than Gemini 3.5 Flash at 9.00 dollars. On input, MiniMax at 0.30 dollars is exactly five times cheaper than Gemini at 1.50 dollars. Those are big multiples, and at scale they decide budgets outright. If your workload is dominated by output tokens, which most generation and agentic workloads are, MiniMax is the dramatically cheaper option token for token.

The nuance that changes the picture is MiniMax's 512K threshold. The 0.30 and 1.20 dollar rates apply only while a prompt stays at or below 512K tokens. Cross that line, and MiniMax M3 supports prompts up to 1,048,576 tokens, and the rate doubles to 0.60 dollars input and 2.40 dollars output for that request. At that point MiniMax is still cheaper than Gemini, but the output advantage falls from seven and a half times to a little under four times. Gemini 3.5 Flash, by contrast, charges a single flat 1.50 dollars input and 9.00 dollars output across its entire 1,048,576-token window, with no step-up and cached input priced far lower still. So the honest way to compare is by workload: for short-to-medium prompts MiniMax is far cheaper, and for prompts that consistently run near the context ceiling the gap narrows and Gemini's flat pricing becomes a real predictability advantage.

Capability and Benchmarks

Benchmarks are a minefield when vendors pick favorable evaluations and report them their own way, so we discipline this hard: we lean on the one independent evaluator that scores both models with the same battery, Artificial Analysis, and we treat any vendor-reported figure as an attributed claim, not a verified fact. That distinction is the backbone of this section.

The cleanest signal here is the Artificial Analysis Intelligence Index, because it is one evaluator running the same battery on both models: Gemini 3.5 Flash at 50 against MiniMax M3 at 44, a clear six-point lead for Gemini. That is the strongest independent evidence in the matchup, and it points to Gemini 3.5 Flash as the more capable model on measured general intelligence. It is also an impressive score for a tier engineered around speed rather than maximum quality, and MiniMax score of 44 is impressive in turn for an open-weight model you can carry off and run yourself.

We keep independent scores and vendor-reported scores strictly apart, because mixing them is how misleading comparisons get built. Beyond the independent index, MiniMax publishes its own benchmark results for MiniMax M3, and the one most often quoted is a 59 percent result on the SWE-bench Pro agentic coding benchmark. We present that strictly as a vendor self-reported number run on MiniMax's own harness: it is not charted by an independent evaluator, and there is no verified head-to-head to build from it, so it cannot be lined up against the independent index as if it were the same kind of evidence. Treated honestly, it tells you MiniMax is competitive on agentic coding by its own measurement, and nothing more precise than that. The number we trust for a like-for-like read remains the Artificial Analysis Intelligence Index, and it favors Gemini.

Speed and Latency

Raw intelligence is only half the story for a Flash-class model, and it is where Gemini 3.5 Flash makes its strongest case. Google positions the Flash tier around throughput and low latency, and it is engineered to run several times faster than frontier-class models while holding frontier-adjacent quality. In practice that speed changes what the model is good for: it is the natural choice for interactive assistants, autocomplete and inline suggestions, real-time agents, and any high-volume pipeline where you are paying in wall-clock time as much as in tokens. A model that answers in a fraction of the time can also be called more often inside an agent loop for the same latency budget, which compounds its usefulness.

MiniMax M3 is efficient rather than fast-first. Its sparse mixture-of-experts design, activating about 23 billion of 428 billion parameters per token, keeps serving costs low, which is what enables the aggressive pricing, but low serving cost is not the same as low latency, and MiniMax does not market M3 as a speed leader. For batch and cost-sensitive work that is fine, and often ideal. For interactive, latency-bound experiences, Gemini 3.5 Flash has the clearer advantage, and it is a genuine differentiator rather than a rounding error.

Context Window and Handling

On raw context length the two models are tied: both carry a 1,048,576-token window, roughly one million tokens, which is plenty for large codebases, long documents, and multi-file agent tasks. Neither has an advantage on how much you can stuff into a single prompt.

The difference, again, is economic. Gemini 3.5 Flash prices the whole window at one flat rate, so a 900K-token prompt costs the same per token as a 9K-token prompt. MiniMax M3 splits its window at 512K: below that line it is dramatically cheaper, and above it the rate doubles. That makes MiniMax the better deal for the large majority of prompts, which sit well under 512K tokens, and Gemini the more predictable and eventually cheaper-relative choice for workloads that routinely push toward the ceiling. If you are architecting a system around consistently huge contexts, model MiniMax at its doubled rate, not its headline one, and weigh that against Gemini's flat number.

Architecture and Deployment: Open vs Closed

It is tempting to treat two API endpoints as interchangeable, but the engineering and licensing underneath shape cost, control, and where each can run. The two could hardly be more different in philosophy.

Gemini 3.5 Flash is a closed model, so Google discloses behavior rather than internals. What you get is a managed service: the model runs on Google's infrastructure, you reach it through the Gemini API, Google AI Studio, or the Gemini app, and you never see or move the weights. The upside is operational simplicity and consistency, no hardware to provision, no serving stack to maintain, a single flat price across the full context window, and the speed and reliability of Google's platform. The trade-offs are the usual closed-model ones: no self-hosting, no downloadable weights, and no path to run the model inside your own boundary for data sovereignty.

MiniMax M3 is the opposite, transparent at the weight level because the model ships openly. It is a mixture-of-experts design with 428 billion total parameters and roughly 23 billion active per token, so only a fraction of the network fires for any given token; that is how MiniMax serves a frontier-adjacent model at budget prices. It pairs that with its MSA sparse attention scheme to keep the 1,048,576-token context affordable, and it is natively multimodal. Because the weights are open, you can download the model, run it on your own GPUs, fine-tune it, and keep every token inside your own infrastructure. The cost of that freedom is operational: a 428-billion-parameter model needs real GPU capacity to serve, so self-hosting shifts spend from per-token billing to hardware and engineering.

The practical upshot is that Gemini 3.5 Flash gives you a polished, fast, managed, flat-priced model you cannot inspect or move, while MiniMax M3 gives you an inspectable, movable model that you can operate yourself at a much lower token price. Neither philosophy is wrong; they serve different cost, control, and sovereignty profiles.

Total Cost of Ownership

Per-token price is the headline, but the real economics depend on prompt length, volume, and whether you self-host. Here is how to think about it without overstating the case in either direction.

For the hosted-API path on short-to-medium prompts, MiniMax M3 is decisively cheaper: seven and a half times cheaper on output and five times cheaper on input than Gemini 3.5 Flash. A pipeline that generates, say, a billion output tokens a month of mostly short prompts costs about 9,000 dollars on Gemini and about 1,200 dollars on MiniMax M3 at standard rates, a large and repeatable saving. Two things temper that. First, if your prompts routinely exceed 512K tokens, MiniMax's rate doubles and the same billion output tokens costs about 2,400 dollars, so the saving against Gemini's flat 9,000 dollars narrows to a little under four times rather than seven and a half. Second, what you buy for Gemini's premium is not just tokens: it is the faster responses and the higher independent capability score, which for latency-bound or quality-bound work can be worth the difference.

For the self-hosted path, the calculus flips from per-token billing to capital and operations. MiniMax M3's open weights remove the API meter entirely, but you pay in hardware: a 428-billion-parameter mixture-of-experts model needs substantial GPU capacity and a serving stack, plus the engineering to run it reliably. For a team with steady, high volume and the operational maturity to run model infrastructure, self-hosting MiniMax M3 can be the cheapest option of all, and the only one that guarantees data never leaves your premises. For a team with spiky or modest volume, the hosted MiniMax API is the sensible comparison, and there it is straightforwardly far cheaper than Gemini on standard-length prompts. Gemini 3.5 Flash earns its premium on speed, managed reliability, and the higher capability score, not on cheaper tokens.

How We Tested

Honesty about methodology matters more in a close, cross-vendor comparison than almost anywhere else. Here is exactly what is hands-on and what is research.

We ran both models side by side through their APIs on reasoning and coding prompts to confirm they behave as documented: Gemini 3.5 Flash's managed endpoint, its speed, and its flat-rate behavior, and MiniMax M3's endpoint, multimodal handling, and the 512K pricing threshold. Those behavioral observations are first-hand. What we have not done is stand up a self-hosted MiniMax M3 cluster or run weeks of controlled, identical-task benchmarking of the two against each other on a private suite. For that reason, every capability claim that rests on a number is attributed to its source: the Artificial Analysis Intelligence Index version 4.1 for the one independent, same-evaluator read, Google for its own reported figures on Gemini, and MiniMax for its own self-reported figures, each labeled as such and never stacked against an independent score as if they were equivalent evidence. We took all pricing directly from each vendor rather than trusting secondhand summaries. Where we could not verify a like-for-like number, most importantly on agentic coding, where only MiniMax reports a figure and it is self-reported, we said so and left the head-to-head uncommitted. That is the standard we hold ourselves to, and it is the only honest way to compare a fast closed model against an open-weight one.

Winner by Category

A single overall winner would be dishonest here, because these two models are tuned for different buyers. Here is who wins what.

• Best for measured intelligence: Gemini 3.5 Flash. It sits at 50 on the Artificial Analysis Intelligence Index version 4.1, a clear six points ahead of MiniMax M3 at 44.
• Best for speed: Gemini 3.5 Flash. Engineered to run several times faster than frontier-class models, it is the clear choice for interactive and latency-sensitive work.
• Best for cost: MiniMax M3. Exactly seven and a half times cheaper per output token and exactly five times cheaper per input token at standard rates, before you even consider self-hosting.
• Best for open weights and self-hosting: MiniMax M3. Downloadable open weights you can run, fine-tune, and keep inside your own infrastructure; Gemini cannot be self-hosted at all.
• Best for predictable long-context pricing: Gemini 3.5 Flash. Both hold a 1,048,576-token window, but Gemini is flat across it while MiniMax doubles above 512K tokens.
• Best for managed simplicity: Gemini 3.5 Flash. A hosted Google endpoint with nothing to provision or operate, against an open model you either rent from MiniMax or run yourself.

Pros and Cons

When to Pick Each

When to pick Gemini 3.5 Flash

Pick Gemini 3.5 Flash when speed and measured capability matter more than squeezing the last cent out of per-token cost. If you want the higher independent score, it leads the Artificial Analysis Intelligence Index version 4.1 at 50 to 44, and it delivers that at high throughput on Google's infrastructure with nothing for you to run. Pick it when latency is part of the product: for interactive assistants, real-time agents, and autocomplete, its speed is a genuine differentiator. Pick it when your prompts are consistently long, because its flat 1.50 dollars input and 9.00 dollars output rate holds across the entire 1,048,576-token window, so you never hit the step-up that MiniMax applies above 512K tokens. And pick it if you already live inside the Gemini API, AI Studio, or the Gemini app and want a fast, reliable model that drops straight into that stack. For a team that values speed, capability, and managed simplicity over the absolute lowest token price, Gemini 3.5 Flash is the natural default.

When to pick MiniMax M3

Pick MiniMax M3 when cost or control dominate. If you are running high-volume inference on short-to-medium prompts where token spend is the binding constraint, being seven and a half times cheaper on output and five times cheaper on input changes what is economically viable. Pick it if you need to own your weights: the open release lets you self-host, fine-tune, and keep data entirely inside your own infrastructure, which no closed model can offer. Pick it if native multimodal handling on a budget is central to your workflow. Just size your budget on the rate you will actually pay: if most prompts sit under 512K tokens, MiniMax is dramatically cheaper, and if they routinely run longer, model the doubled rate before you commit. You give up a slice of independent capability and the speed edge, but you get most of the quality at a fraction of the price, plus the freedom to run the model yourself.

Final Verdict

This is a split verdict by use case, tilted toward Gemini 3.5 Flash on intelligence and speed and toward MiniMax M3 on cost and openness. On the one independent signal that scores both the same way, the Artificial Analysis Intelligence Index version 4.1, Gemini leads 50 to 44, a clear six-point edge, and it delivers that capability several times faster than frontier-class models. Both models share the same 1,048,576-token context window, but Gemini prices it at a single flat rate where MiniMax M3 doubles above 512K tokens. MiniMax M3, in return, costs exactly seven and a half times less per output token and exactly five times less per input token at standard rates, ships open weights you can self-host and fine-tune, and is natively multimodal, a genuinely strong package for a budget open model.

We did not crown a single overall winner because the two are not really competing for the same buyer. If you want the higher independent capability score, the fastest responses, or flat and predictable long-context pricing, the answer is Gemini 3.5 Flash. If you are cost-constrained or want to own and self-host your weights, the answer is MiniMax M3. Both answers are correct, for different people. Every capability figure here is drawn from the Artificial Analysis independent index or explicitly labeled as a vendor self-reported claim, and only the pricing is taken directly from each vendor.

If you are weighing these two against the wider field, we compared a closed frontier tier against another open-weight budget challenger in Claude Sonnet 5 vs DeepSeek V4 and GPT-5.5 vs DeepSeek V4, and pitted the best coding score against the cheapest open model in GLM-5.2 vs DeepSeek V4. For the deep dive on each model on its own, see our full Gemini 3.5 Flash review and MiniMax M3 review, and for where they land in the broader field, our best AI coding tools of 2026.

Frequently Asked Questions

Is Gemini 3.5 Flash better than MiniMax M3?

On independent measured intelligence, yes, by a clear margin. Gemini 3.5 Flash scores 50 on the Artificial Analysis Intelligence Index version 4.1, while MiniMax M3 scores 44 on the same independent index, a six-point lead. Gemini is also built for speed, running several times faster than frontier-class models, and it charges a single flat rate across its full context window. But MiniMax M3 is far cheaper, five times cheaper on input and seven and a half times cheaper on output at standard rates, and it ships open weights you can self-host. So the better model depends on whether you optimize for intelligence and speed or for cost and control. This is a split verdict, not a knockout.

How much cheaper is MiniMax M3 than Gemini 3.5 Flash?

At standard rates, MiniMax M3 costs 0.30 dollars per million input tokens and 1.20 dollars per million output tokens, against Gemini 3.5 Flash at 1.50 dollars input and 9.00 dollars output. That makes MiniMax exactly five times cheaper on input and exactly seven and a half times cheaper on output. The catch is that MiniMax standard pricing only holds for prompts up to 512K tokens; above that threshold its rate doubles to 0.60 dollars input and 2.40 dollars output, which narrows the output gap. Gemini charges a single flat rate across its entire context window. All prices were taken directly from each vendor.

Does MiniMax M3 pricing really double above 512K tokens?

Yes. MiniMax M3 bills 0.30 dollars per million input tokens and 1.20 dollars per million output tokens only while the prompt stays at or below 512K tokens. Once a request crosses 512K tokens, and MiniMax M3 supports up to 1,048,576, the per-token rate doubles to 0.60 dollars input and 2.40 dollars output for that request. It is still cheaper than Gemini 3.5 Flash at that point, but the discount shrinks. Gemini, by contrast, charges 1.50 dollars input and 9.00 dollars output flat across its full 1,048,576-token window, so for genuinely long prompts the price gap narrows and Gemini flat pricing becomes more attractive.

Is MiniMax M3 open source?

MiniMax M3 is open weight rather than fully open source in the strictest sense. It is a mixture-of-experts model with 428 billion total parameters and about 23 billion active per token, and MiniMax publishes the weights so you can download, self-host, fine-tune, and run the model on your own hardware. That is the decisive difference from Gemini 3.5 Flash, which is a closed model you can only reach through Google. If owning and controlling the model matters to you, MiniMax M3 is the only one of the two that offers it.

Can I self-host Gemini 3.5 Flash or MiniMax M3?

You can self-host MiniMax M3 because its open weights are downloadable, so you can run it inside your own infrastructure for data control and to remove per-token billing. You cannot self-host Gemini 3.5 Flash, which is available only through Google as a managed service, namely the Gemini API, Google AI Studio, and the Gemini app. Running MiniMax M3 yourself is not free, though: a 428-billion-parameter mixture-of-experts model needs serious GPU capacity, so self-hosting trades per-token cost for hardware and operational cost.

Which model is smarter, Gemini 3.5 Flash or MiniMax M3?

On the one independent evaluator that scores both models the same way, Gemini 3.5 Flash is the smarter model. It scores 50 on the Artificial Analysis Intelligence Index version 4.1, six points ahead of MiniMax M3 at 44. That composite blends reasoning, knowledge, math, and coding evaluations into a single number, and it is the cleanest like-for-like signal available for this matchup. The gap is clear but not enormous, and MiniMax score of 44 is a strong result for an open-weight model you can run yourself.

Which model is faster?

Gemini 3.5 Flash. Speed is its defining feature: Google built the Flash tier for high throughput and low latency, and it runs several times faster than frontier-class models while holding frontier-adjacent intelligence. MiniMax M3 keeps its own inference costs low with a sparse mixture-of-experts design, which is efficient, but raw speed and latency are not its headline claim the way they are for Gemini 3.5 Flash. For interactive or latency-sensitive workloads, Gemini 3.5 Flash is the stronger choice on responsiveness.

What is the context window for each model?

Both models offer a 1,048,576-token context window, roughly one million tokens, so they are effectively tied on raw length. The more important difference is pricing behavior across that window. Gemini 3.5 Flash charges one flat rate for the whole window, while MiniMax M3 doubles its rate for any prompt above 512K tokens. So while the two are matched on how much context they can hold, Gemini is the more predictable choice for consistently long-context work, and MiniMax is the cheaper choice as long as prompts stay under the 512K threshold.

Which model is better for coding?

There is no independent head-to-head coding score in this matchup, so we are careful here. On overall independent capability, the Artificial Analysis Intelligence Index favors Gemini 3.5 Flash, which is the cleanest like-for-like signal available. MiniMax reports strong agentic coding results on its own harness, but those are vendor self-reported figures rather than independently charted ones, so we treat them as claims rather than scoreboard entries. For most coding buyers the practical decision is cost against capability and speed: MiniMax M3 is far cheaper and self-hostable, while Gemini 3.5 Flash is stronger on the one independent index and much faster.

Is Gemini 3.5 Flash the same as the Gemini 3 Flash Preview?

No. Gemini 3.5 Flash, launched at Google I/O on May 19, 2026, is a distinct, more capable, and more expensive model than the earlier Gemini 3 Flash Preview. The Preview is a separate, lower-cost, lower-capability tier; Gemini 3.5 Flash is the generally available fast tier with frontier-adjacent intelligence, a 1,048,576-token context window, and pricing of 1.50 dollars per million input tokens and 9.00 dollars per million output tokens. If you are comparing pricing or benchmarks, make sure the figures refer to Gemini 3.5 Flash and not the cheaper Preview, because they are easy to confuse.

Which should I choose for a high-volume production workload?

For pure high volume where token spend is the binding constraint, MiniMax M3 is usually the rational default, because being five times cheaper on input and seven and a half times cheaper on output changes what is economically viable at scale, and self-hosting the open weights can cut cost further. Choose Gemini 3.5 Flash when you want the higher independent capability score, the fastest responses, managed reliability on Google infrastructure, and a flat rate that does not double on long prompts. Many teams run both: Gemini for the calls that need speed or the extra capability, MiniMax M3 for the cheap bulk.

When were these models released and is this comparison current?

Gemini 3.5 Flash was released at Google I/O on May 19, 2026. MiniMax M3 was released on June 1, 2026. This comparison was last updated in July 2026, with pricing taken directly from each vendor and the independent capability figures drawn from the Artificial Analysis Intelligence Index version 4.1. Any vendor-reported benchmark is labeled as such and kept separate from independent scores.