MiniMax M2.5

Name: MiniMax M2.5
Author: MiniMax

Multimodal

MiniMax

MiniMax M2.5 is an intermediate update to the MiniMax model family, offering improvements in reasoning accuracy, coding quality, and instruction adherence over M2.1. It supports multimodal inputs and extended context, with strong benchmark results across diverse tasks.

Key Specifications

Parameters

230.0B

Context

Release Date

February 11, 2026

Average Score

66.7%

API Documentation Model Weights Results Blog

Timeline

Key dates in the model's history

Announcement

February 11, 2026

Last Update

February 17, 2026

Today

March 25, 2026

Technical Specifications

Parameters

230.0B

Training Tokens

Knowledge Cutoff

Family

Capabilities

MultimodalZeroEval

Benchmark Results

Model performance metrics across various tests and benchmarks

Programming

Programming skills tests

SWE-Bench Verified

• Self-reported

80.0%

Other Tests

Specialized benchmarks

BFCL_v3_MultiTurn

• Self-reported

77.0%

BrowseComp

Standard evaluation • Self-reported

76.0%

MEWC

• Self-reported

74.0%

GDPval-MM

• Self-reported

59.0%

SWE-Bench Pro

• Self-reported

55.0%

Multi-SWE-Bench

• Self-reported

46.0%

License & Metadata

License

mit

Announcement Date

February 11, 2026

Last Updated

February 17, 2026

Compare MiniMax M2.5

All comparisons

vs MiniMax M2.1 vs MiniMax M2.7 vs Mercury 2 vs GPT-5.3 Codex vs GLM-4.7-Flash vs GPT-5.2 Codex vs GLM-5 vs Claude Sonnet 4.6

Articles about MiniMax M2.5

MiniMax M2.7 Will Go Open Weights Within Two Weeks

MiniMax head of engineering confirms M2.7, the self-evolving AI model, will be released as open weights. Here's what that means for developers.

March 24, 2026

2 min

Similar Models

All Models

MiniMax M2

MiniMax

230.0B

Best score:0.8 (GPQA)

Released:Oct 2025

Price:$1.00/1M tokens

Pixtral Large

Mistral AI

MM124.0B

Released:Nov 2024

Price:$2.00/1M tokens

Step-3.5-Flash

StepFun

MM196.0B

Best score:0.9 (TAU)

Released:Feb 2026

Price:$0.10/1M tokens

Llama 4 Maverick

Mistral Large 3 (675B Instruct 2512)

Mistral AI

MM675.0B

Best score:0.4 (GPQA)

Released:Dec 2025

Price:$0.50/1M tokens

GLM-4.5V

Zhipu AI

MM108.0B

Released:Aug 2025

Price:$0.60/1M tokens

Llama 4 Scout

K-EXAONE-236B-A23B

LG AI Research

236.0B

Released:Dec 2025

Price:$0.60/1M tokens

Recommendations are based on similarity of characteristics: developer organization, multimodality, parameter size, and benchmark performance. Choose a model to compare or go to the full catalog to browse all available AI models.