Agent Mag
SUBSCRIBE
MiMo-V2.5-Pro
All models
XiaomiXiaomiMiMoReleased 2026-04-22

MiMo-V2.5-Pro

1.0M context$1.00/M input$3.00/M output

MiMo-V2.5-Pro is Xiaomi's flagship AI model designed for high-performance agentic tasks, complex software engineering, and long-horizon operations. It features a context length of up to 1 million tokens, enabling integration with diverse agent frameworks and handling tasks requiring extensive tool usage. The model excels in benchmarks such as ClawEval, GDPVal, and SWE-bench Pro, showcasing its ability to autonomously complete professional tasks over extended periods.

What is MiMo-V2.5-Pro?

MiMo-V2.5-Pro is an AI model from Xiaomi that Agent Mag tracks for pricing, context window, modalities, benchmarks, and API compatibility. Builders can use this page to compare MiMo-V2.5-Pro against other models for agent workflows and production deployments.

Model ID

MiMo-V2.5-Pro is Xiaomi's flagship AI model designed for high-performance agentic tasks, complex software engineering, and long-horizon operations. It features a context length of up to 1 million tokens, enabling integration with diverse agent frameworks and handling tasks requiring extensive tool usage. The model excels in benchmarks such as ClawEval, GDPVal, and SWE-bench Pro, showcasing its ability to autonomously complete professional tasks over extended periods.

Modalities
Input
text
Output
text
Supported Parameters
frequency_penaltyinclude_reasoningmax_tokenspresence_penaltyreasoningresponse_formatstoptemperaturetool_choicetoolstop_p
Strengths
  • Handles long-horizon tasks with up to 1M context length
  • Excels in benchmarks like ClawEval and SWE-bench Pro
  • Autonomously completes professional tasks requiring extensive tool usage
  • Optimized for integration with agent frameworks
  • Strong performance in complex software engineering
Limitations
  • High hallucination rate in knowledge-based tasks (75.5%)
  • Low accuracy in omniscience-related benchmarks (22.6%)
  • Limited performance in research-level physics reasoning (4.0%)
Recommended Use Cases
Integration with agent frameworks for long-horizon tasks
Complex software engineering and coding workflows
Autonomous completion of professional tasks
Handling economically valuable tasks
Scientific reasoning and instruction-following scenarios

More from Xiaomi

MiMo-V2.5
1.0M ctx$0.400/M

MiMo-V2.5 is a native omnimodal model by Xiaomi. It delivers Pro-level agentic performance at roughly half the inference cost, while surpassing MiMo-V2-Omni in multimodal perception across image and video understanding...

MiMo-V2-Omni
262K ctx$0.400/M

MiMo-V2-Omni is a frontier omni-modal model that natively processes image, video, and audio inputs within a unified architecture. It combines strong multimodal perception with agentic capability - visual grounding, multi-step...

MiMo-V2-Pro
1.0M ctx$1.00/M

MiMo-V2-Pro is Xiaomi's flagship foundation model, featuring over 1T total parameters and a 1M context length, deeply optimized for agentic scenarios. It is highly adaptable to general agent frameworks like...

MiMo-V2-Flash
262K ctx$0.090/M

MiMo-V2-Flash is an open-source foundation language model developed by Xiaomi. It is a Mixture-of-Experts model with 309B total parameters and 15B active parameters, adopting hybrid attention architecture. MiMo-V2-Flash supports a...

Related content

Browse the AI models directory

Compare pricing, local installs, context windows, and modality filters across the full model catalog.

Explore AI agent tools

Find frameworks, SDKs, and infrastructure tools that pair with this model in production workflows.

Read Agent Mag model and agent articles

See Agent Mag coverage of model benchmarks, agent frameworks, and deployment patterns.

Data enriched Apr 24, 2026. Pricing from OpenRouter API.