minimax M3

What minimax M3 is for

minimax M3 is a reasoning model: it spends output tokens working through a problem before answering, which makes it slower per useful word and better on problems that need the working shown. minimax released it in 2026-06 with a single 230B size under minimax License (open-weight), with a 128K context window. Budget for the extra tokens the thinking phase consumes. The catalogue records its strength as latest minimax M-series (June 2026).

Sizes and memory

The 230B needs about 130 GB of weights at Q4_K_M. That is the whole range; there is no larger or smaller variant in this record. Recommended VRAM adds roughly half again on top of the weights for the KV cache and the runtime.

Size	Weights at Q4_K_M	Recommended VRAM	Smallest card in our guide that holds it
230B	130 GB	195 GB	Apple M3 Ultra

Hardware that runs minimax M3

2 of the 19 cards in our hardware guide hold at least one size of minimax M3 at Q4_K_M. The cheapest that does is the Apple M3 Ultra at about $4000, running the 230B.

The largest size any of them holds is the 230B at 130 GB, on the Apple M3 Ultra (192 GB at 800 GB/s). Bandwidth, not capacity, sets the speed once a size fits, so a card further down the list with faster memory will generate more tokens per second on the same weights than a larger, slower one.

What people use it for

The catalogue lists 3 recorded uses for minimax M3: coding, agentic, professional productivity. It is tagged chat, tools, coding, agentic, thinking.

Where minimax M3 sits in the minimax family

Our catalogue holds 5 minimax entries. Ordered by release date, minimax M3 (2026-06) is the 5th of them, and the most recent. minimax M2 and minimax M2.1 and minimax M2.5 carry the same recorded memory footprint, so choosing between those and this one is a question of behaviour rather than of hardware: minimax M3 is the one recorded here for latest minimax M-series (June 2026).

It ships under the minimax License (open-weight), which is neither a standard open-source licence nor a closed one. Read the terms before shipping anything commercial on top of it.

Context window: 128K

128K tokens is roughly 96,000 words at the usual ratio, which is a small repository or a whole book. Two cautions come with a window this size: the KV cache at full length can rival the weights for memory, and models rarely use the far end of their advertised context as well as they use the near end. Treat it as headroom, not as a promise.

How to run minimax M3 locally

Our catalogue does not record registry tags, so look the current tag up in the Ollama library or on Hugging Face before pasting these. The commands below are the shape of the workflow, not a copy-and-paste recipe.

Option 1: Ollama (simplest)

# find the tag for minimax M3 at ollama.com/library
ollama run <tag>

Option 2: Mullama (production)

mullama pull <tag>
mullama run <tag>

Option 3: llama.cpp (CLI)

# download a GGUF from Hugging Face, searching for "minimax M3 GGUF"
./llama-cli -m model.Q4_K_M.gguf -p "Hello, AI!"

Option 4: Python with Mullama or llama-cpp-python

from mullama import Model, Context
model = Model.load("model.Q4_K_M.gguf", n_gpu_layers=99)
ctx = Context(model, n_ctx=4096)
print(ctx.generate("Hello, AI!", 256))

Sources

Model record from src/data/models.json (2026-06-29); card capacities from src/data/gpus.json (2026-06-29). Upstream: Curated from ollama.com/library + community benchmarks + paperswithcode + huggingface.