How much does MiniMax M2.5 cost?

MiniMax M2.5 costs $0.27/M input tokens and $0.95/M output tokens, with separate rates for Cache Read at $0.03/M tokens.

What is the context length of MiniMax M2.5?

MiniMax M2.5 has a 198,000 token context window. It supports up to 32,768 completion tokens.

Does MiniMax M2.5 support tool calling and structured outputs?

Yes. MiniMax M2.5 accepts tools and tool_choice for function calling. It also supports structured outputs via a JSON schema in response_format.

Which providers serve MiniMax M2.5?

MiniMax M2.5 is served by 4 providers on OpenRouter: Venice, AtlasCloud, GMICloud, and DigitalOcean. Requests are routed to the best available provider, with automatic failover to the others, and you can pin or exclude providers with provider routing.

When was MiniMax M2.5 released?

MiniMax M2.5 was released on February 12, 2026.

MiniMax: MiniMax M2.5

minimax/minimax-m2.5

Model weights

Compare

MiniMax-M2.5 is a SOTA large language model designed for real-world productivity. Trained in a diverse range of complex real-world digital working environments, M2.5 builds upon the coding expertise of M2.1 to extend into general office work, reaching fluency in generating and operating Word, Excel, and Powerpoint files, context switching between diverse software environments, and working across different agent and human teams. Scoring 80.2% on SWE-Bench Verified, 51.3% on Multi-SWE-Bench, and 76.3% on BrowseComp, M2.5 is also more token efficient than previous generations, having been trained to optimize its actions and output through planning.

Modalities

In / out price

$0.27 / $0.95per 1M

Context

198K

Released

Feb 12, 2026

Providers

Different companies host the same model. OpenRouter routes your request to one of them based on the routing mode you pick: Balanced (price + speed), Nitro (fastest), or Floor (cheapest).

Latency / throughput


$0.27	$0.95	$0.03	1.11s	58 tok/s	97.29%
$0.295	$1.20	$0.06	2.32s	53 tok/s	99.78%
$0.30	$1.20	$0.03	1.08s	39 tok/s	99.48%
$0.30	$1.20	$0.06	1.04s	58 tok/s	100.00%

Input $0.27Output $0.95
58 tok/s97.29% uptime
Input $0.295Output $1.20
53 tok/s99.78% uptime
Input $0.30Output $1.20
39 tok/s99.48% uptime
Input $0.30Output $1.20
58 tok/s100.00% uptime

Pricing

The average price customers actually pay for this model, next to the prices providers post. Caching and discounts mean the price actually paid is often well below the listed one.

Weighted average

Input

$0.1141/M tokens

Output

$1.12/M tokens

MiniMax M2.5 — Price history

Price history

No effective price data for this selection

$0.110	$1.199	78.6%	56.6%
$0.116	$0.949	64.1%	26.6%
$0.121	$1.199	74.4%	9.2%
$0.127	$1.199	63.9%	7.5%

Effective in $0.110Effective out $1.199
Token share
56.6%
Effective in $0.116Effective out $0.949
Token share
26.6%
Effective in $0.121Effective out $1.199
Token share
9.2%
Effective in $0.127Effective out $1.199
Token share
7.5%

Uptime

Uptime is the percentage of the past 3 days that at least one provider was responding to requests. Availability is the percentage of time that inference was successfully served. OpenRouter continuously monitors and uses the next-best provider when one returns an error.

Uptime (3d)

100.00%

Availability (3d)

99.81%

Availability over the last 3 days

Last 72 hours

Availability 99.81%

3 Days Ago2 Days AgoYesterdayNow

Availability over the last 24 hours

OpenRouter Availability

99.56%

Without Routing

99.03%

When an error occurs in an upstream provider, we can recover by routing to another healthy provider, if your request filters allow it. You can access per-provider uptime data programmatically through the Endpoints API. Learn more about our load balancing and provider routing options.

Benchmarks

Scores on standardized evaluations and this model’s own outputs on shared benchmark prompts. Higher percentages are better — and each rank shows where this model lands among the models evaluated on that source.

Benchmark score summary for MiniMax: MiniMax M2.5 (Artificial Analysis, Design Arena and Vals AI)
Source	Benchmark	Score
Artificial Analysis	MiniMax-M2.5 GPQA Diamond	84.8%
Artificial Analysis	MiniMax-M2.5 HLE	20.5%
Artificial Analysis	MiniMax-M2.5 IFBench	71.6%
Artificial Analysis	MiniMax-M2.5 τ²-Bench Telecom	95.3%
Artificial Analysis	MiniMax-M2.5 AA-LCR	73.3%
Artificial Analysis	MiniMax-M2.5 CritPt	1.1%
Artificial Analysis	MiniMax-M2.5 Terminal-Bench Hard	34.8%
Design Arena	MiniMax M2.5 Models Arena 3D Elo	1149
Design Arena	MiniMax M2.5 Models Arena Code Categories Elo	1196
Design Arena	MiniMax M2.5 Models Arena Data Visualization Elo	1168
Design Arena	MiniMax M2.5 Models Arena Game Development Elo	1166
Design Arena	MiniMax M2.5 Models Arena SVG Elo	1138
Design Arena	MiniMax M2.5 Models Arena UI Component Elo	1163
Design Arena	MiniMax M2.5 Models Arena Website Elo	1214
Vals AI	MiniMax-M2.5 AIME	88.8%
Vals AI	MiniMax-M2.5 CaseLaw v2	53.5%
Vals AI	MiniMax-M2.5 CorpFin v2	59.6%
Vals AI	MiniMax-M2.5 Finance Agent v1.1	38.6%
Vals AI	MiniMax-M2.5 GPQA Diamond	82.1%
Vals AI	MiniMax-M2.5 IOI v1	6.7%
Vals AI	MiniMax-M2.5 LiveCodeBench	79.2%
Vals AI	MiniMax-M2.5 LegalBench	80.0%
Vals AI	MiniMax-M2.5 MedQA	92.5%
Vals AI	MiniMax-M2.5 MMLU Pro	80.1%
Vals AI	MiniMax-M2.5 ProofBench v1	4.0%
Vals AI	MiniMax-M2.5 SWE-bench	74.2%
Vals AI	MiniMax-M2.5 TaxEval v2	68.2%
Vals AI	MiniMax-M2.5 Terminal-Bench 2.0	41.6%
Vals AI	MiniMax-M2.5 Vibe Code Bench v1.1	14.9%

Apps

Public apps that send the most traffic to this model. Good signal for what real production workloads look like — and a hint at which use cases this model is best suited for.

OpenClaw

OpenClaw is an open-source AI agent that connects to your messaging apps and takes real actions on your behalf, from running commands and browsing the web to managing files and sending emails.

11.3Btokens

Hermes Agent

Hermes Agent is an open-source, self-improving AI agent by Nous Research that runs persistently with memory across sessions, and builds reusable skills from experience. It comes with 40+ built-in tools, including web search, browser automation, and vision, plus scheduled automations and subagents.

3.62Btokens

Cline

Cline is an open-source AI coding agent that lives inside your IDE, autonomously exploring your codebase, editing files, running terminal commands, and using browser automation.

641Mtokens

Service AI Multimodal Gateway

new

275Mtokens

There are many coding agents, but this one is yours.

177Mtokens

Top 5 Apps

111M

Rest of OpenRouter

58.1M

Not currently available through EU in-region routing. In-region routing docs

More models from MiniMax

MiniMax H3 Image Highspeed

MiniMax H3 Image Highspeed is the faster version of MiniMax's H3 Image model. It takes the same inputs and limits as MiniMax H3 Image: it creates images from text prompts and edits or composes images from 1 to 16 reference images, with output up to 20 megapixels and aspect ratios from 1:3 to 3:1.

Image$3 / $7.60

MiniMax H3 Image

MiniMax H3 Image is MiniMax's image generation and editing model. It creates images from text prompts and edits or composes images from 1 to 16 reference images, with output up to 20 megapixels and aspect ratios from 1:3 to 3:1.

Image$1.50 / $3.80

MiniMax M3.1 Flash

MiniMax-M3.1-Flash is a multimodal model from MiniMax. Compared with M3, it delivers significant improvements across multiple benchmarks, particularly in complex agentic tasks, coding, and cybersecurity. It is built for complex, multi-step co-work automation, understanding real codebases, fixing bugs, and building features, and visual understanding and creation, including frontend and presentation creation. It is an open-weight mixture-of-experts model with 23B active / 428B total parameters.

Text1.0M context$0.075 / $0.30

H3 Max

MiniMax H3 Max is a video-generation model from MiniMax, jointly released with fal.ai. Derived through additional training from MiniMax H3, it is designed for faster text-to-video and image-to-video generation with controlled first-frame or last-frame keyframes.

Videofrom $0.05/second

MiniMax H3 is a lightweight, open-weights video generation model from MiniMax. It is designed for precise multimodal editing and controlled content generation, including instruction-guided edits, text and brand rendering, and video-to-video motion transfer.

The model is suited for commercial creative workflows across advertising, e-commerce, gaming, and interface design, with native audiovisual output for reference-driven generation.

Videofrom $0.13/second

Speech 2.8 HD

MiniMax Speech 2.8 HD is a text-to-speech model from MiniMax. It is suited for applications that generate spoken audio from text and accepts arbitrary MiniMax voice IDs.

Speech$100/M characters

Speech 2.8 Turbo

MiniMax Speech 2.8 Turbo is a text-to-speech model from MiniMax. It is suited for applications that generate spoken audio from text and accepts arbitrary MiniMax voice IDs.

Speech$60/M characters

MiniMax M3

MiniMax-M3 is a multimodal foundation model from MiniMax. It supports text, image, and video inputs with text output, a 1M-token context window, and is suited for long-horizon agentic work, coding, and tool use. It is built on MiniMax Sparse Attention (MSA), which replaces full attention with KV-block selection to cut per-token compute at long context — roughly 1/20 the cost of the previous generation at 1M tokens, with substantially faster prefill and decode while retaining quality across most tasks.

Trained as a native multimodal model on interleaved data and tuned for multi-turn, production-like collaboration via an interactive user-simulator framework, the model is oriented toward sustained, multi-step tasks rather than single-turn execution.

Text1.0M context$0.23 / $0.96

MiniMax M3 (batch)

Text524K context$0.224 / $0.88

Hailuo 2.3

Hailuo 2.3 is a video generation model from MiniMax. It accepts text prompts and reference images as input and generates video output, supporting both text-to-video and image-to-video workflows. It is suited for creative content production, cinematic scene generation, and character animation, with a focus on realistic motion and expressive character rendering.

Video$0.0817/second

MiniMax M2.7

MiniMax-M2.7 is a next-generation large language model designed for autonomous, real-world productivity and continuous improvement. Built to actively participate in its own evolution, M2.7 integrates advanced agentic capabilities through multi-agent collaboration, enabling it to plan, execute, and refine complex tasks across dynamic environments.

Trained for production-grade performance, M2.7 handles workflows such as live debugging, root cause analysis, financial modeling, and full document generation across Word, Excel, and PowerPoint. It delivers strong results on benchmarks including 56.2% on SWE-Pro and 57.0% on Terminal Bench 2, while achieving a 1495 ELO on GDPval-AA, setting a new standard for multi-agent systems operating in real-world digital workflows.

Text197K context$0.21 / $0.84

MiniMax M2-her

MiniMax M2-her is a dialogue-first large language model built for immersive roleplay, character-driven chat, and expressive multi-turn conversations. Designed to stay consistent in tone and personality, it supports rich message roles (user_system, group, sample_message_user, sample_message_ai) and can learn from example dialogue to better match the style and pacing of your scenario, making it a strong choice for storytelling, companions, and conversational experiences where natural flow and vivid interaction matter most.

Text66K context

MiniMax M2.1

MiniMax-M2.1 is a lightweight, state-of-the-art large language model optimized for coding, agentic workflows, and modern application development. With only 10 billion activated parameters, it delivers a major jump in real-world capability while maintaining exceptional latency, scalability, and cost efficiency.

Compared to its predecessor, M2.1 delivers cleaner, more concise outputs and faster perceived response times. It shows leading multilingual coding performance across major systems and application languages, achieving 49.4% on Multi-SWE-Bench and 72.5% on SWE-Bench Multilingual, and serves as a versatile agent “brain” for IDEs, coding tools, and general-purpose assistance.

To avoid degrading this model's performance, MiniMax highly recommends preserving reasoning between turns. Learn more about using reasoning_details to pass back reasoning in our docs.

Text205K context

MiniMax M2

MiniMax-M2 is a compact, high-efficiency large language model optimized for end-to-end coding and agentic workflows. With 10 billion activated parameters (230 billion total), it delivers near-frontier intelligence across general reasoning, tool use, and multi-step task execution while maintaining low latency and deployment efficiency.

The model excels in code generation, multi-file editing, compile-run-fix loops, and test-validated repair, showing strong results on SWE-Bench Verified, Multi-SWE-Bench, and Terminal-Bench. It also performs competitively in agentic evaluations such as BrowseComp and GAIA, effectively handling long-horizon planning, retrieval, and recovery from execution errors.

Benchmarked by Artificial Analysis, MiniMax-M2 ranks among the top open-source models for composite intelligence, spanning mathematics, science, and instruction-following. Its small activation footprint enables fast inference, high concurrency, and improved unit economics, making it well-suited for large-scale agents, developer assistants, and reasoning-driven applications that require responsiveness and cost efficiency.

To avoid degrading this model's performance, MiniMax highly recommends preserving reasoning between turns. Learn more about using reasoning_details to pass back reasoning in our docs.

Text205K context

MiniMax M1

MiniMax-M1 is a large-scale, open-weight reasoning model designed for extended context and high-efficiency inference. It leverages a hybrid Mixture-of-Experts (MoE) architecture paired with a custom "lightning attention" mechanism, allowing it to process long sequences—up to 1 million tokens—while maintaining competitive FLOP efficiency. With 456 billion total parameters and 45.9B active per token, this variant is optimized for complex, multi-step reasoning tasks.

Trained via a custom reinforcement learning pipeline (CISPO), M1 excels in long-context understanding, software engineering, agentic tool use, and mathematical reasoning. Benchmarks show strong performance across FullStackBench, SWE-bench, MATH, GPQA, and TAU-Bench, often outperforming other open models like DeepSeek R1 and Qwen3-235B.

Text1M context

MiniMax-01

MiniMax-01 is a combines MiniMax-Text-01 for text generation and MiniMax-VL-01 for image understanding. It has 456 billion parameters, with 45.9 billion parameters activated per inference, and can handle a context of up to 4 million tokens.

The text model adopts a hybrid architecture that combines Lightning Attention, Softmax Attention, and Mixture-of-Experts (MoE). The image model adopts the “ViT-MLP-LLM” framework and is trained on top of the text model.

To read more about the release, see: https://www.minimaxi.com/en/news/minimax-01-series-2

Text1M context

MiniMax: MiniMax M2.5

Providers