MiniMax-M2.5
MiniMaxMiniMaxOpen WeightMIT · Usage Commercial
Description
MiniMax M2.5 is the world's first production-level model designed natively for Agent scenarios. Building on the M2.1 foundation, M2.5 delivers significant improvements in programming, tool calling, search, and office productivity. With only 10B activation parameters from its 230B MoE architecture, it achieves competitive performance against top international models like Claude Opus 4.6 while maintaining high throughput and efficient inference. M2.5 supports full-stack development for PC, App, and cross-platform applications, and excels in agentic workflows including automated customer support, data-analysis pipelines, and complex task execution.
Date de sortie
2026-02-12
Paramètres
230.0B
Longueur du contexte
205K
Modalités
image, text
Radar de capacités
31
general
43
coding
85
reasoning
57
science
70
agents
60
multimodal
Classements
| Domaine | #Rang | Score | Source |
|---|---|---|---|
| Capacité agentique | 69 | 47.0 | LS |
| Classement codage | 113 | 70.0 | AA |
| Classement général | 89 | 72.0 | AA |
| Science | 121 | 65.0 | AA |
Scores de benchmarks (LLM Stats)
(LLM Stats (zeroeval))Agents
BrowseCompOpenAI (2025)
76.3%Aut.
MEWC
74.4%Aut.
SWE-Bench ProPrinceton NLP (2024)
55.4%Aut.
VIBE-Pro
54.2%Aut.
Code
SWE-Bench Verified
80.2%Aut.
Multi-SWE-Bench
51.3%Aut.
Finance
GDPval-MM
59.0%Aut.
General
BFCL_v3_MultiTurn
76.8%Aut.
Indices d'évaluation AA
(Artificial Analysis)Intelligence Index(Artificial Analysis)34.5
Tau2(Sierra + U Toronto + Vector Institute (2025))1.0
Gpqa(NYU + Cohere + Anthropic (2023))0.8
Lcr(Artificial Analysis)0.7
Ifbench(Google Research (2023))0.7
Scicode(UIUC + Argonne National Lab (2024))0.4
Terminalbench Hard(Stanford × Laude Institute (2026))0.3
Hle(Center for AI Safety + Scale AI (2025))0.2
Scores par catégorie LLM Stats
(LLM Stats (zeroeval))Search80
Frontend Development80
Agents70
Multimodal60
Reasoning60
Finance60
General60
Code60
Tarification
Prix d'entrée$0.3 / 1M tokens
Prix de sortie$1.2 / 1M tokens
Prix mixte (3:1)$0.525 / 1M tokens
Prix de lecture cache$0.03 / 1M tokens
Prix d'écriture cache$0.375 / 1M tokens
Vitesse
Tokens/sec0.0
Délai du premier token0.00s
Temps de réponse0.00s
Classement des Prix par Fournisseur
Classement des Prix par Fournisseur
15 fournisseurs
Moins cher: MiniMaxPlus cher: TensorX
FournisseurEntréeSortie
1MiniMaxMoins cher
$0
$0
2OpenRouter
$0.22
$0.9
3NanoGPT
$0.3
$1.2
4Alibaba (China)
$0.3
$1.2
5ZenMux
$0.3
$1.2
6NovitaAI
$0.3
$1.2
7Kilo Gateway
$0.3
$1.2
8FrogBot
$0.3
$1.2
9Vercel AI Gateway
$0.3
$1.2
10MiniMax (minimax.io)
$0.3
$1.2
11HPC-AI
$0.3
$1.2
12MiniMax (minimaxi.com)
$0.3
$1.2
13OrcaRouter
$0.3
$1.2
14Merge Gateway
$0.3
$1.2
15TensorX
$0.3
$1.2
Comparer les prix entre différents fournisseurs API pour ce modèle.