> MiniMax-M2.5 is an open-weights AI model from MiniMax, released 12 Feb 2026. Index score 146 (GPT-4 = 100), rank 65 of 229 ranked models. API price $0.27 in / $1.08 out per million tokens, 205K context.

[Models](https://themodelindex.org/models/) /[MiniMax](https://themodelindex.org/labs/minimax/)

# MiniMax-M2.5

MiniMax

Released 12 Feb 2026

Open

Date verified

[Compare](https://themodelindex.org/compare/minimax-m2-5-vs-gpt-5-4/) [Announcement](https://huggingface.co/MiniMaxAI/MiniMax-M2.5)

146135–158

Index score

#65 of 229 models

2/ 8

Core tests taken

2 of 6 areas · plus 4 anchor tests

−24

Versus the frontier at release

best then: GPT-5.3-Codex

Replaced by [MiniMax-M2.7](https://themodelindex.org/models/minimax-m2-7/)

$0.27/ $1.08

Price per million tokens

input / output, list price

205K

Context window

tokens it reads at once

## Where it sits · on the frontier

_Chart: Model Index score of every scored model since late 2022, with the best closed and best open model over time_

o3 reached this score 10 months earlier.

## Test results · by area · tick = best by any model

Show 2 anchor tests

Reasoning & knowledge

no evidence yet

GPQA Diamond

no result

Humanity's Last Exam

no result

Maths

evidence

FrontierMath T1-3

no result

Coding

evidence

SWE-bench Verified

no result

Terminal-Bench

43%

Agents

no evidence yet

APEX-Agents

no result

Novel problems

evidence

ARC-AGI-2

5%

Long tasks

no evidence yet

METR time horizon

no result

Independent results from the Epoch AI benchmarking hub, best reasoning setting. The core tests are shown here; anchor tests also inform the score and can be shown above.

## What others say · rank on each leaderboard

The Model Index

independent tests, our method

#65

of 219 · 146

Epoch Capabilities Index

overall capability from many benchmarks

#66

of 239 · 146.7

LMArena · Coding

coding questions in chat

#140

of 383 · 1446

LMArena · Creative writing

creative writing

#151

of 386 · 1357

LMArena · Text

overall chat quality, judged by people

#152

of 388 · 1391

LMArena · Maths

maths questions in chat

#150

of 373 · 1393

LMArena · WebDev

building web apps and front ends

#83

of 123 · 1386

Further left is better. Leaderboards measure different things: LMArena is people voting blind between two answers, the others are test scores. Click any row for the source.

## Best at · top three among current models

Replaced by MiniMax-M2.7; it no longer leads any category among current models.

Where this model places in the top three of today's models, on category leaderboards and on our core tests.

## Where to use it · prices checked 1 Oct 2026

#### From the lab

MiniMax Agent

app · subscription

↗

MiniMax API

API

↗

#### API providers · 7 · per million tokens, in / out

V

Venice

$0.27 / $0.95

cheapest

S

StreamLake

$0.27 / $1.08

↗

A

AtlasCloud

$0.29 / $1.2 · fp8

↗

D

DigitalOcean

$0.30 / $1.2

↗

F

Friendli

$0.30 / $1.2

↗

N

Novita

$0.30 / $1.2 · fp8

↗

M

Minimax

$0.60 / $2.4 · fp8

↗

#### Run it yourself

🤗

Official weights

MiniMaxAI/MiniMax-M2.5

↗

Provider prices come from OpenRouter's live listing; going direct to a provider can differ. Subscriptions are the lab's own apps.

## Facts and sources

Released

12 Feb 2026

Announcement

[huggingface.co ↗](https://huggingface.co/MiniMaxAI/MiniMax-M2.5)

Release entry

MiniMax-M2.5

Scored as

MiniMax-M2.5

Artificial Analysis index

–

Output speed

–

Hugging Face

[MiniMaxAI/MiniMax-M2.5 ↗](https://huggingface.co/MiniMaxAI/MiniMax-M2.5)

## MiniMax releases · around this one

MiniMax-M2

Open

27 OCT 2025

MiniMax-M2.1

Open

23 DEC 2025

MiniMax-M2.5

Open

12 FEB 2026

index 146

MiniMax-M2.7

Open

18 MAR 2026

index 143

MiniMax-M3

Open

1 JUN 2026

index 145

---
Source: https://themodelindex.org/models/minimax-m2-5/ · The Model Index · data as of 2026-10-01