LLM Reference

MiniMax M2.5 Highspeed

Released
2026-02-12
Last refreshed
2026-06-29
Status
Researched 80d ago
Open sourceCommercial use: permittedCodingRAGAgentsLong contextClassificationJSON / Tool use

MiniMax M2.5 Highspeed is worth evaluating for coding, rag, and agents when its provider route and context window match the workload.

Use it for

  • Teams evaluating coding, rag, and agents
  • Workloads that can use a 205k context window
  • Buyers comparing 3 tracked provider routes

Do not use it for

  • Vision or document-understanding workloads
Specifications
Released
2026-02-12
Context
205k
Max output
131,072
Parameters
230B (10B active)
Architecture
Decoder Only
Specialization
general
Openness
Open source
License
MITOSI-approvedCommercial use: permitted
Weights
Unknown
Code
Unknown
Created by

Developing AI for gaming and entertainment.

Minhang, Shanghai, China
Founded 2021
Website
Pricing
Output / 1M
$2.40
Input / 1M
$0.600

Cheapest of 3 routes · Novita AI

About

MiniMax M2.5 Highspeed is MiniMax's inference-optimized variant of M2.5, released simultaneously in February 2026. It delivers identical intelligence and outputs to standard M2.5 through a specialized inference engine at lower latency. The model supports a 204,800-token context window, 131,072-token max output, function calling, structured output, and reasoning. API model ID: MiniMax-M2.5-highspeed. It is designed for latency-sensitive interactive applications and automated agent pipelines.

MiniMax M2.5 Highspeed is an open-source model in the MiniMax M2 family. The structured metadata tracks a 205k-token context window, reasoning, function calling, tool use, and structured outputs. This page tracks provider routes through MiniMax, Vercel AI Gateway, and Novita AI, with the cheapest tracked route listed at $0.6 input and $2.4 output per 1M tokens. Headline tracked benchmarks include SWE-bench Verified 80.2.

Top use-case fit: coding, agents, and build tasks

Coding

Q/$ C

1 relevant benchmark in the decision map.

RAG

Included by capability and metadata signals in the decision map.

Agents

Q/$ C

1 relevant benchmark in the decision map.

Provider price ladder

Compare all 3

Compare API pricing across 3 providers for input and output tokens, batch, and cached reads when available.

ProviderInput / 1MOutput / 1MCacheRoute
Novita AI$0.600$2.40-
Serverless
Vercel AI Gateway$0.600$2.40read $0.030
Serverless
MiniMax---
ServerlessPartial

Capabilities

ReasoningFunction CallingTool UseStructured Outputs

Benchmark peer barsfor Coding

Benchmark scores(1)

Scores are benchmark-specific and are direction-aware: the same numeric gap can mean very different outcomes across suites. Use the leaderboard context and this model's provider route to decide whether the winning margin is meaningful for your workload.
BenchmarkScoreVersionEvaluationSource
SWE-bench Verified80.2M2 (resolved%)Observed 2026-06-07Source

Migration checks

No linked migration route is available for this model yet.

Frequently asked questions

What is the context window of MiniMax M2.5 Highspeed?

MiniMax M2.5 Highspeed has a context window of 205k tokens.

What is the max output of MiniMax M2.5 Highspeed?

MiniMax M2.5 Highspeed can generate up to 131,072 output tokens.

How much does MiniMax M2.5 Highspeed cost?

MiniMax M2.5 Highspeed is available at $0.6/1M input tokens through Vercel AI Gateway.

When was MiniMax M2.5 Highspeed released?

MiniMax M2.5 Highspeed was released on 2026-02-12.

Which providers offer MiniMax M2.5 Highspeed?

MiniMax M2.5 Highspeed is available from 3 providers: MiniMax, Vercel AI Gateway, Novita AI.

What benchmarks has MiniMax M2.5 Highspeed been tested on?

MiniMax M2.5 Highspeed has been evaluated on 1 benchmark, including SWE-bench Verified.