MiniMax M3 model profile

Profile facts

Profile type

Model research profile

Provider or publisher

MiniMax

Context window

1,048,576 tokens

Snapshot date

2026-09-06

Knowledge cutoff

Not disclosed

Reference catalog ID

minimax/minimax-m3

When to pick MiniMax M3

MiniMax M3 is a reference profile: MiniMax publishes it, Kendr documents it for comparison, and it carries no Kendr alias to call.

Each condition below is derived from this model's own values in the 2026-09-06 snapshot, compared only against models publishing the same field.

• Reach for it when a whole corpus has to fit in one prompt. The 1.05M-token context window is in the top quarter of the 163 models publishing a limit here, so long documents can go in whole rather than chunked.
• Reach for it when the input includes images. The snapshot records image input, so screenshots, scans, and diagrams can be sent directly instead of described.
• Reach for it when the model has to call tools. Tool calling is recorded in the snapshot, so this route can drive an agent loop rather than only answer in prose.
• Look elsewhere when you need to call it through Kendr today. This is a research reference profile with no Kendr alias. The facts and evidence here are published for comparison; the model is not routable on Kendr.
• Look elsewhere when the task is hard reasoning. Its 35.7 Artificial Analysis intelligence index is below the median of the 32 models carrying that field. Scores from different suites are not interchangeable, so treat this as one signal rather than a ranking.

Context, modalities, and identity

MiniMax M3 accepts text, image, and video and returns text. The snapshot records prompt caching, reasoning controls, structured output, text generation, tool calling, video input, and vision as capabilities or interfaces.

Provider or publisher

MiniMax

Context window

1,048,576 tokens

Maximum output

512,000 tokens

Input modalities

text, image, and video

Output modalities

text

Knowledge cutoff

Not disclosed

Model overview

MiniMax-M3 is a multimodal foundation model from MiniMax. It supports text, image, and video inputs with text output, a 1M-token context window, and is suited for long-horizon agentic work, coding, and tool use. It is built on MiniMax Sparse Attention (MSA), which replaces full attention with KV-block selection to cut per-token compute at long context — roughly 1/20 the cost of the previous generation at 1M tokens, with substantially faster prefill and decode while retaining quality across most tasks. Trained as a native multimodal model on interleaved data and tuned for multi-turn, production-like collaboration via an interactive user-simulator framework, the model is oriented toward sustained, multi-step tasks rather than single-turn execution.

Reference model ID

minimax/minimax-m3

Canonical version

minimax/minimax-m3-20260531

Hugging Face ID

MiniMaxAI/Minimax-M3

Capabilities and supported controls

Capabilities and parameters are reported from the dated catalog and source records; their presence does not guarantee identical behavior across every provider route.

Supported parameters

frequency_penalty, include_reasoning, logit_bias, logprobs, max_tokens, min_p, presence_penalty, reasoning, repetition_penalty, response_format, seed, stop, structured_outputs, temperature, tool_choice, tools, top_k, top_logprobs, top_p

• prompt caching
• reasoning controls
• structured output
• text generation
• tool calling
• video input
• vision

Dated pricing snapshot

The dated reference snapshot lists $0.23 input and $0.96 output per million tokens. Cached input is $0.05 per million tokens.

These are dated third-party reference prices, not Kendr prices or an availability offer. Routes, tiers, caching, region, and provider policy can change the landed price.

Price date

2026-09-06

Input

$0.23 per 1M tokens

Cached input

$0.05 per 1M tokens

Output

$0.96 per 1M tokens

Provider routes, performance, and uptime

The snapshot retains 12 provider endpoints. Provider prices, context limits, p50 performance, and uptime can differ by route and are not Kendr guarantees.

Active provider endpoints

12

Best p50 latency

0.39 s

Best p50 throughput

150 tok/s

Availability with routing

99.89% over the sampled window

Availability without routing

97.84% over the sampled window

Performance date

2026-09-06

Benchmark evidence and limitations

The 2026-09-06 snapshot includes 3 Artificial Analysis indexes, 15 Design Arena categories, and AutoExacto results for 14 provider observations. Publisher-reported launch results are labeled and are not presented as Kendr measurements; scores from different suites are not treated as interchangeable.

Third-party catalog fields normalized from the cited model dataset; benchmark methodology and coverage differ by source.

Intelligence Index

35.7

Coding Index

58.6

Agentic Index

31

AutoExacto coverage

14 provider observations over 32 days

AutoExacto provider benchmarks

Rolling provider observations from the cited third-party benchmark view. The lookback is 32 days; GPQA Diamond and TAU-Bench Airline measure different abilities and should not be blended into one score.

Popularity and market context

MiniMax M3 ranked #16 in the cited trailing-7-day third-party catalog usage dataset as of 2026-09-06. Observed within the cited trailing-7-day third-party catalog usage dataset; this is not global AI market share.

Third-party catalog rank

#16

Global market share

Not inferred

Observation date

2026-09-06

Frequently asked questions

Sources and evidence dates

1. MiniMax M3 third-party catalog record (catalog, checked 2026-09-06)
2. Third-party model catalog methodology (methodology, checked 2026-09-06)
3. MiniMax official model documentation (primary, checked 2026-09-06)
4. Artificial Analysis capability indices methodology (benchmark, checked 2026-09-06)
5. Design Arena leaderboard and methodology (benchmark, checked 2026-09-06)
6. MiniMax M3 third-party catalog record (benchmark, checked 2026-09-06)