MiniMax M2.5

by MiniMax

MiniMax M2.5 — multimodal frontier model strong on long-context and agentic workloads.

ToolsStructured output
Context window
200K
Max output
197K
Input price
$0.30 /1M
Output price
$1.10 /1M

Benchmarks

Independent benchmark scores — composite indices for reasoning, coding, and math, plus individual eval scores where available.

Global rank#62 of 505 LLMs
TierStrong
Output speed67 tok/s
First token1.67s
Intelligence Index41.9
Coding Index37.4
GPQA Diamond
85%
Humanity's Last Exam
19%
SciCode
43%
IFBench
72%
Long-context reasoning
66%
Terminal-Bench Hard
35%
τ²-Bench Telecom
95%

Available routes

The same MiniMax M2.5 runs on 4 different routes through the Opper gateway. Each route has its own data-handling posture.

ProviderRegionZero data retentionTrainingLoggingContractInputOutput
InceptronEUOpt-inNo
7–30 days default with per-workspace overrides
PAYG$0.30$1.10
InfercomEUNo
Pending verification
PAYG$0.32$1.30
NebiusUSOpt-inNo
Per privacy policy: not detailed.
PAYG$0.30$1.20
NovitaUSNone
Not published; appendix available upon request.
PAYG$0.30$1.20

Human review: Abuse-flagged only

Other models from MiniMax

Start building with 300+ models

One API key. Every major provider. Up and running in minutes.

Get startedView Documentation
MiniMax M2.5 by MiniMax — pricing, benchmarks, hosting routes | Opper AI