MiniMax M3
9051°
<minimax-m3>
Apps that support ModelTrends tags let you run/install models by copying this code in your prompt with no setup. Look for our icon in apps that support it, like Prism (below).
Run this model
Offers / Ads
Easy Creator (Prism) Web
Unlimited · flat fee
Best for beginners. Describe what you want in a chat and the picture comes back, no learning curve.
PirateDiffusion Telegram
Unlimited · flat fee
Images and video from a chat, with short commands. The most convenient tool for Telegram users, with a web companion.
Pro studio (SpyGlass) Web
Unlimited · flat fee
Best for power users. Full control over models, settings and workflows in a browser studio.
Discord bot (Vector)
Unlimited · flat fee
Best for communities. Add it to your own server and run any workflow from a channel.
Unified Editor Web
Unlimited · flat fee
One screen for every model and workflow: prompts, references, edits and video, built into your Graydient account.
OpenRouter Terminal / API
Pay per token
Chat with this model in the browser or call it from your code with one API key.
Local apps
Requires setup
Set up local programs if your computer has capable hardware.
ModelDuel Web
Pay per use
Generate the same prompt on up to 50 checkpoint/base models at a time.
Guides for this model
How to choose an AI model in October 2026: the closed source news, then the open weights worth downloading
Best local LLM by use case (October 2026): the right open model for every job
How much VRAM do I actually need for local AI?
Open weights vs open source vs API-only: what you are actually allowed to do
How to Install Ollama and Run a Local AI (Windows, Mac, Linux)
MiniMax | 1,048,576 token context | $0.30/M input, $1.20/M output. 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.
Captain's notes
Notes from 2026-10-13
MiniMax M3 is a 428B-parameter sparse mixture-of-experts model from MiniMax, activating roughly 23B parameters per token, with a 1-million-token context window and native support for text, image, and video inputs. Its MiniMax Sparse Attention architecture cuts per-token compute to about 1/20th of the previous generation at full context, with 9× faster prefill and 15× faster decode. Weights are open, the API is cheap, and the design targets sustained agentic work rather than one-shot queries.
Known for
- 1M-token context at a fraction of the compute cost of dense attention models
- Native multimodal training across text, image, and video from the first step
- 9× prefill and 15× decode speedups over M2 at maximum context
- Three switchable reasoning modes: always-on, adaptive, or disabled
- Tuned for multi-turn agentic tasks and tool use, not just single-turn prompts
Watch out for
- At 428B total parameters, local deployment demands serious hardware; check VRAM requirements before downloading.
- Open weights, but verify the licence terms for your use case before building a commercial product on top.
- Video and image input quality at extreme context lengths is not independently benchmarked here.
Prompts to try
Multi-file refactor task
You are working on a Python web scraper split across three files: fetcher.py handles HTTP requests with retries, parser.py extracts structured data using BeautifulSoup, and main.py orchestrates the pipeline. The scraper silently drops items when a page returns a 429. Diagnose the likely cause, propose a fix with exponential backoff, and rewrite the relevant sections of fetcher.py and main.py to implement it.
Long document Q&A
I am going to paste a long contract below. Read the entire document, then answer these three questions: 1) What are the termination conditions for either party? 2) Are there any automatic renewal clauses, and if so what notice period applies? 3) Which party bears liability for data breaches? Cite the clause number for each answer. [Paste contract text here]
Image-grounded analysis
Attached is a screenshot of a dashboard. Identify every metric visible, note which ones are outside a normal operating range based on what you can see, and suggest the two most likely root causes for any anomalies. Then write a three-sentence summary a non-technical manager could read.
Compare with
- MiniMax M1API-only model1,000,000 token context · $0.55/M input, $2.20/M output
- MiniMax M2.5Open-weights model204,800 token context · $0.27/M input, $0.95/M output
- MiniMax-01Open-weights model1,000,192 token context · $0.20/M input, $1.10/M output
- MiniMax M2.1Open-weights model204,800 token context · $0.30/M input, $1.20/M output
More text models from minimax
Questions people ask
- Can I run MiniMax M3 locally?
- Yes, the weights are open and hosted at MiniMaxAI/MiniMax-M3 on Hugging Face. MiniMax supports SGLang, vLLM, Transformers, KTransformers, unsloth, and ATOM as inference frameworks, though at 428B total parameters you will need substantial hardware.
- What does MiniMax M3 cost via API?
- The API is priced at $0.30 per million input tokens and $1.20 per million output tokens, with a 1,048,576-token context window.
- How is M3 different from M2?
- M3 replaces full attention with MiniMax Sparse Attention, delivering 9× faster prefill and 15× faster decode at 1M context while reducing per-token compute to roughly 1/20th of M2; it also adds native video input and a configurable reasoning mode that M2 did not expose.
Captain's verdictA rare open-weights model that takes the million-token context window seriously enough to rebuild the attention mechanism around it.
Model card on Hugging Face
Captain writes these notes from the model's own description and model card. Prices and context sizes are the provider's at the time of writing and can change.
Ranking badge
Updated Oct 22, 2026
Made this model? Show this badge on your page.
Paste it into your Hugging Face card or Civitai description.
Also ranked: #19 Hottest Text / LLM model · #19 Hottest Text / LLM text model
Not this creator?
Get your own badge
MiniMax M3 Popularity Trend
Heat score since 31 May 2026
Heat score
9,051°
This week
▲
29.0%
9,017° then, 9,051° now
This month
▲
231%
8,892° then, 9,051° now
Past year
n/a
Wasn't released/tracked
7,000°
Floor
The scale runs from 0° to 9,999°. Everything below 7,000° is kept for models no source reports on yet, which carry a stand-in number there so lists stay complete. 7,000° is where measured data begins: a model with real data but almost no downloads sits at the floor, and the measurements fill the degrees above it. The floor says nothing about the model; only the degrees above it do, and they are what separates one model from another.
+1,727°
OpenRouter tokens
5,730,756,128,120 tokens this model processed through OpenRouter in the last 30 days, prompts and answers together. Every 10,000,000 tokens count as one download in the download term, 573,075 in all; the last day's tokens join the adoption rate. Source: OpenRouter (openrouter.ai/rankings).
+324°
Adoption rate
About 20,825 downloads a day, measured between two readings of the sources, never guessed from a lifetime average. Speed weighs a quarter of the download term, so a fresh surge shows without letting a one-day spike outrank years of use.
Show detailed breakdown
Every number on this site can be checked. This model's heat score is 9,051°, and here it is built one ingredient at a time, with the running total after each: the floor, the degrees each source adds, the bonus for the kind of model and what the age of its weights takes away. The colors are the bands of the chart above. The glossary at the end defines every term.
- 1Start at the floor.Why 7,000° and not zero: the scale runs from 0° to 9,999°, and the part below 7,000° is reserved for models no source reports on yet, which carry a stand-in number there so the lists stay complete and never outrank a model with data. 7,000° is where measured data begins. A model with real data but almost no downloads sits here; everything that follows is measurement, and the room above the floor holds ten tenfold steps in downloads at 300° each. The floor is the same for every model, so it decides nothing; the degrees above it do.7,000°running total
- 2Add the downloads.573,075 download-equivalents in all, from OpenRouter tokens. Ten times the downloads adds the same 300°, so size matters but never runs away with the score.+1,727°8,727°running total
- 3Add the adoption rate.About 20,825 downloads a day right now, measured between two readings, at a quarter of the weight.+324°9,051°running total
- =The heat score.Where this model stands on every list of this site. Nothing else adjusts it.9,051°
Glossary: the terms used here
- Heat score
- One number for how much a model is used and wanted right now, from 0° to 9,999°. Every list on this site that says hottest is sorted by it and nothing else.
- Floor
- The 7,000° every model with real data starts from. The scale runs from 0° to 9,999°; the part below the floor is reserved for stand-ins (models without data), so that a model with data always ranks above one without. The floor is the same for everyone and decides nothing; the degrees above it are the measurement.
- Download-equivalents
- Everything the score treats as a download, added up: downloads themselves, each torrent seeder as 1,000, each AI Horde image and each run on a generation site as one.
- Log scale
- Counting by tens, not by ones: going from 1,000 to 10,000 downloads adds the same degrees as going from 10,000 to 100,000. It keeps giants from crowding out everything else.
- Likes
- Thumbs up and likes from every source, worth half a download each on the log scale.
- Adoption rate
- How many downloads a day a model gets right now, from the difference between two readings of a source. A quarter of the weight of the download term.
- Measured and estimated
- Measured means two readings of a source exist and the difference is real. Estimated means only today's total is known and the past is spread over the model's life; the site says which one it is showing.
- Seeder
- Someone who keeps a copy of the weights online for others to download over a torrent, or a hosted mirror doing the same. Each one counts as 1,000 downloads.
- AI Horde
- A volunteer network that generates images for free on donated computers. The number of images it has made with a model is public, and each counts as one download.
- Generation sites
- Sites that host their own copies of a model and let people generate with it, such as PixAI, SeaArt and Yodayo. Their downloads count like Civitai's and each generation run as one download.
- OpenRouter tokens
- A text model served through an API, such as Claude or GPT, has no weights to download, so its usage is measured in tokens: the pieces of text it read and wrote through OpenRouter, a gateway many apps use, over the last 30 days. Every 10,000,000 tokens count as one download. Source: OpenRouter (openrouter.ai/rankings), CC BY 4.0.
- Full model (checkpoint)
- The weights themselves, the file that generates on its own. It gets a bonus because add-ons depend on it.
- LoRA and embedding
- Small add-on files that steer someone else's weights toward a style, a character or a concept. They cannot run alone.
- Family and weights
- The base a model is built on, such as SD 1.5, SDXL, Flux or Wan. Every model inherits the release date of its family's weights.
- Age
- Degrees taken off for how long ago the family's weights came out: 150° a year, up to 750°. It keeps old downloads that had years to pile up from freezing the top of the list.
- Family boost
- A small, bounded nudge for LoRAs of a family this site is pushing forward right now, always listed in the breakdown.
- Stand-in score
- A stable number below the floor given to a model no source reports on yet, so lists stay complete. It never ranks above a model with data.
- New! (grace period)
- A model less than 7 days old shows New! instead of a score and holds no place in the rankings. Its score is computed from day one, but a model one day old grows faster than anything, so without the wait the week's #1 would always be whatever came out yesterday rather than what held up for a week.
- Trend
- How a model's growth over a period compares with the typical model's growth over the same period. Faster than typical reads as a rise, slower as a fall.
- Typical model
- The middle of the catalog: half the models with a figure grew faster over the period, half slower.
How this model's popularity has moved against the rest of the catalog: its growth in downloads, likes and usage over each period, measured against the typical model's growth over the same period, with its heat score at the time. Above the typical growth reads as a rise, below it as a fall. The past is reconstructed from today's totals, the model's first day and its measured download speed where two readings exist; a model that did not exist at the start of a period shows n/a.
MiniMax M3 is a text model for the Text / LLM family listed on ModelTrends.ai, a read-only catalog of open source AI image, video and text models. Compare it with other Text / LLM models, check its heat score to see how it is trending, and open it on PirateDiffusion or BitVector to try it.
Browse by Model Family
Text / LLM Models
434
Anima Models
528
Chroma Models
19
Fibo Models
2
Flux Models
1,496
Flux 2 / Klein Models
1,239
Flux 3 Models
3
Moss Models
3
Iris Models
1
MiniMax H3 Models
364
Hunyuan Models
204
Ideogram Models
20
Krea2 Models
717
Qwen Models
20
Qwen2 Models
47
Ltx2 Models
153