MiMo V2.5 Pro
MiMo V2.5 Pro is capable in long context, instruction following, factuality, and coding; and behind the leaders in reasoning and agentic tasks. Too few results yet to rate safety, math, multimodal tasks, or multilingual tasks.
Capability profile
Bars show the model's median result as a share of the leading model's, per capability. Select a row to see its results.
Too few results yet to rate Safety, Math, Multimodal or Multilingual.
Price
$0.43input$0.87outputper million tokens
From Artificial Analysis · 3 providers tracked · All prices
Evidence
72results on53benchmarks
- 2 independently verified
- 35 aggregator
- 35 vendor-reported
From 6 sources · latest Oct 8, 2026 · How verification works
API features
Tool callingStructured outputsReasoning
As listed by OpenRouter
Research
5 papers reference MiMo V2.5 ProMiMo V2.5 Pro benchmark results
72 results on 53 benchmarks, grouped by capability. Each bar is the result as a share of the capability leader's; every result links to its source.
11.7% behind the leader1 of 3 ranked benchmarks measured
- 79.67Oct 8, 2026
Show 1 more long context resultHide 1 long context result
- 41.67Oct 8, 2026
20.6% behind the leader1 of 3 ranked benchmarks measured
- IFBench79.86Oct 8, 2026aa_ifbench
Show 1 more instruction following resultHide 1 instruction following result
- IFBench42.72Oct 8, 2026aa_ifbench
22.9% behind the leader2 of 4 ranked benchmarks measured
- AA-Omniscience · Non-hallucination75.30Oct 8, 2026omniscienceNonHallucination
- AA-Omniscience · Accuracy22.42Oct 8, 2026omniscienceAccuracy
Show 2 more factuality resultsHide 2 factuality results
- AA-Omniscience · Accuracy27.18Oct 8, 2026omniscienceAccuracy
- AA-Omniscience · Non-hallucination10.76Oct 8, 2026omniscienceNonHallucination
23.4% behind the leader7 of 10 ranked benchmarks measured
- 78.90Oct 7, 2026
- SciCode50.58Oct 8, 2026aa_scicode
- Terminal-Bench 2.165.17Oct 8, 2026terminalbenchV21
- 1478.43Sep 24, 2026
- Terminal-Bench Hard43.18Oct 8, 2026aa_terminalbench_hard
- 57.20Oct 7, 2026
- 39.60Oct 7, 2026
Show 3 more coding resultsHide 3 coding results
- SciCode39.12Sep 4, 2026aa_scicode
- 65.20Sep 22, 2026
- Terminal-Bench Hard35.61Oct 8, 2026aa_terminalbench_hard
25.6% behind the leader3 of 6 ranked benchmarks measured
- GPQA Diamond86.57Oct 8, 2026gpqa
- Humanity's Last Exam35.68Oct 8, 2026aa_hle
- 4.00Oct 8, 2026
Show 6 more reasoning resultsHide 6 reasoning results
- 1.14Oct 8, 2026
- GPQA Diamond76.16Oct 8, 2026gpqa
- GPQA Diamond66.70Oct 7, 2026GPQA
- Humanity's Last Exam14.78Oct 8, 2026aa_hle
- 34.00Oct 7, 2026
- 48.00May 3, 2026
44.1% behind the leader4 of 7 ranked benchmarks measured
- 31.17Oct 8, 2026
- τ-Bench V3 · Banking9.90Oct 8, 2026tauBanking
- Terminal-Bench 4.00.00Oct 8, 2026
Show 3 more agentic resultsHide 3 agentic results
- 2.43Oct 8, 2026
- 38.23Oct 8, 2026
- 39.91Jun 15, 2026
More results
Benchmarks outside the capability baskets. They are not ranked against a leader.
- 1468Oct 5, 2026
- τ²-Bench Telecom (AA run)72.51Oct 8, 2026aa_tau2
- -37.80Oct 8, 2026
- AA Intelligence18.29Oct 8, 2026aa_intelligence_index
- τ²-Bench Telecom (AA run)94.15Oct 8, 2026aa_tau2
- AA Intelligence25.99Oct 8, 2026aa_intelligence_index
Show 33 more resultsHide 33 results
- 29.11Jul 7, 2026
- 50.75Jun 18, 2026
- 3.25Oct 8, 2026
- 13.20Sep 22, 2026
- 37.30Oct 7, 2026
- ARC-Challenge97.20Oct 7, 2026ARC-C
- Artificial Analysis Coding Index60.19Jul 7, 2026aa_coding_index
- Artificial Analysis Coding Index36.78Jun 18, 2026aa_coding_index
- 88.40Oct 7, 2026
- 91.50Oct 7, 2026
- Claw Eval (pass@3)64.00Oct 7, 2026Claw-Eval
- 64.00May 15, 2026
- 63.20May 15, 2026
- 90.20Oct 7, 2026
- 40.00Sep 22, 2026
- DeepSWE 1.119.00Sep 22, 2026DeepSWE v1.1
- 86.30Oct 7, 2026
- 41.50Oct 7, 2026
- GDPval-AA 2.11107.00Sep 22, 2026
- 83.60Oct 7, 2026
- 99.60Oct 7, 2026
- 89.80Oct 7, 2026
- 75.60Oct 7, 2026
- 25.00Sep 22, 2026
- 74.10Oct 7, 2026
- 68.50Oct 7, 2026
- 92.80Oct 7, 2026
- 12.50Sep 22, 2026
- 68.40Oct 7, 2026
- 49.10Sep 22, 2026
- 81.30Oct 7, 2026
- 85.60Oct 7, 2026
- τ³-Bench72.90Oct 7, 2026TAU3-Bench
MiMo V2.5 Pro: common questions
Who makes MiMo V2.5 Pro?
MiMo V2.5 Pro is made by Xiaomi.
When was MiMo V2.5 Pro released?
MiMo V2.5 Pro was released on Apr 22, 2026, according to Artificial Analysis.
What is MiMo V2.5 Pro good at?
MiMo V2.5 Pro is capable in long context, instruction following, factuality, and coding; and behind the leaders in reasoning and agentic tasks. Too few results yet to rate safety, math, multimodal tasks, or multilingual tasks.
How much does MiMo V2.5 Pro cost?
MiMo V2.5 Pro costs $0.43 per million input tokens and $0.87 per million output tokens, according to Artificial Analysis. We track its price at 3 providers. At a mix of three input tokens to one output token, it is cheaper than 55% of the 330 priced models we track.
How many benchmarks has MiMo V2.5 Pro been tested on?
We track 72 results for MiMo V2.5 Pro on 53 benchmarks from 6 sources, 2 of them independently verified. The latest was recorded on Oct 8, 2026.
Which API features does MiMo V2.5 Pro support?
OpenRouter lists tool calling, structured outputs, and reasoning for MiMo V2.5 Pro.
About this record
Where MiMo V2.5 Pro's numbers come from, and every name it appears under.
- Tracked since
- Apr 27, 2026
- Newest source mention
- Sep 4, 2026
Where the results come from
Verification: 72 scores · 2 independently verified · 35 aggregator-attributed · 35 vendor-reported. How these tiers are assigned
From 6 sources on 5 sites. Artificial Analysis supplies 35 of them; the 2 independently verified results come from 2 sites. Bars are coloured by trust tier.
- artificialanalysis.ai35
- api.llm-stats.com24
- huggingface.co11
- datasets-server.huggingface.co1
- lmarena.ai1