The LLM benchmark aggregator.

MiMo V2.5 Pro

Xiaomi · 1T · open weights

Strongest in Human preference (#44 of 354), weakest in Jailbreak resistance (#228 of 268). Above par in 11 of 21 scopes. Among the models it meets almost everywhere, it finishes behind Claude Fable 5 and Claude Fable 5.1 and ahead of Gemini 2.5 Pro and NVIDIA Nemotron 3 Ultra 550B A55B.

§ 1 · Profile

What it is good at

Bars run from 50 — the average of the models each source lists — so right of the line is above par. The middle column is the gap to whoever leads that scope.

Human preference63.1−5.7#44/3541/1

Knowledge51.8−10.7#89/1971/3

Academic knowledge54.2−9.7#60/1231/1

Coding46.9−19.3#107/1791/6

Agentic coding46.7−20.2#103/1731/5

Agents51.5−14.9#109/2982/6

Knowledge work52.4−18.3#85/2082/2

Reasoning46.8−18.6#124/1951/5

Science54.5−9#61/1221/1

Mathematics40−23.8#121/1521/2

Professional47.8−14.6#126/1941/2

Finance51−11.4#92/1621/1

Medical46.4−20.1#108/1491/1

Legal44.1−21.8#125/1731/1

Safety49.4−12#198/3291/3

Toxicity avoidance56.4−2.6#59/2681/1

Fairness54.4−15.6#73/2921/2

Secure code57.8−8.4#87/2701/1

Harm refusal46.2−15.3#217/2921/2

Jailbreak resistance36.1−28.9#228/2681/1

§ 2 · Head to head

What it beats, and what beats it

The same models turn up scope after scope. Counted once: where both were placed, who finished higher. Bars run right for MiMo V2.5 Pro, left for the other.

Mistral Small 3.1 24B Instructwon 13 · lost 0

Mistral Medium 3.5won 15 · lost 0

MiMo V2.5won 15 · lost 0

Jamba 1.6 Miniwon 13 · lost 2

Command R+won 13 · lost 2

GPT-4o miniwon 17 · lost 3

Mistral Medium 3.1won 11 · lost 2

Llama 4 Scout Instructwon 16 · lost 3

Ling 3.0 Flashwon 15 · lost 3

GPT-4.1 nanowon 17 · lost 4

Llama 4 Maverick Instructwon 17 · lost 4

Jamba 1.6 Largewon 12 · lost 3

Command Awon 16 · lost 4

Llama 3.3 70B Instructwon 14 · lost 4

Magistral Small 1.2won 13 · lost 4

Gemini 1.5 Prowon 13 · lost 4

DeepSeek V3won 16 · lost 5

Gemini 1.5 Flashwon 12 · lost 4

GPT-5.4 nanowon 12 · lost 4

Mistral Large 3won 12 · lost 4

Gemini 3.1 Flash-Litewon 12 · lost 4

Claude 3.5 Haikuwon 15 · lost 5

Gemma 4 31Bwon 12 · lost 5

DeepSeek R1won 14 · lost 6

Grok 4.1 Fastwon 14 · lost 6

Llama 3.1 8B Instructwon 9 · lost 4

Gemini 2.0 Flashwon 13 · lost 6

Gemini 2.5 Flash-Litewon 10 · lost 5

GPT-5.4 miniwon 10 · lost 5

GLM-4.6won 14 · lost 7

MiniMax M3won 14 · lost 7

GPT OSS 20Bwon 12 · lost 7

Grok 3won 12 · lost 7

Gemini 3.5 Flash-Litewon 9 · lost 6

MiniMax M2.5won 8 · lost 6

Grok 4 Fastwon 8 · lost 6

GPT-4owon 12 · lost 9

O3 Miniwon 12 · lost 9

Grok 4.3won 12 · lost 9

GLM-4.7won 9 · lost 7

GPT-5 nanowon 10 · lost 8

MiniMax M2.7won 11 · lost 9

GLM-4.5won 9 · lost 8

GPT OSS 120Bwon 11 · lost 10

Claude Haiku 4.5won 11 · lost 10

NVIDIA Nemotron 3 Ultra 550B A55Bwon 11 · lost 10

Gemini 2.5 Prowon 11 · lost 10

Qwen3.5 Flashwon 7 · lost 7

GPT-4.1 miniwon 10 · lost 11

GPT-5 miniwon 10 · lost 11

DeepSeek V3.2won 10 · lost 11

Qwen3.6 Pluswon 7 · lost 8

Grok 4.20won 6 · lost 7

Claude 3.7 Sonnetwon 9 · lost 11

Grok 4won 9 · lost 11

Kimi K2won 9 · lost 11

Claude 3.5 Sonnetwon 8 · lost 10

MiMo V2.6 Flashwon 8 · lost 10

Claude Sonnet 4won 9 · lost 12

Kimi K2 Thinkingwon 7 · lost 10

Inkling Smallwon 6 · lost 9

O4 Miniwon 8 · lost 12

Gemini 2.5 Flashwon 8 · lost 12

GPT-4.1won 8 · lost 12

Qwen3 235B A22B Instructwon 8 · lost 12

O1 Miniwon 5 · lost 8

Qwen3 Maxwon 5 · lost 8

Kimi K2.5won 8 · lost 13

Kimi K2.6won 8 · lost 13

O1won 7 · lost 12

MiniMax M2.1won 5 · lost 9

Qwen3.8 27Bwon 5 · lost 10

Inklingwon 7 · lost 14

GPT-5.1won 7 · lost 14

Claude Sonnet 4.5won 7 · lost 14

Qwen3.7 Maxwon 7 · lost 14

Claude Opus 4won 6 · lost 13

GPT-6 Solwon 6 · lost 13

Grok 3 Miniwon 5 · lost 11

GPT-4 Turbowon 4 · lost 9

GLM-5won 4 · lost 10

DeepSeek V4 Flashwon 6 · lost 15

GLM-5.3 Flashwon 6 · lost 15

Gemini 3.5 Flashwon 6 · lost 15

Gemini 3.6 Flashwon 6 · lost 15

Mistral Large 4won 5 · lost 13

Muse Sparkwon 4 · lost 11

GPT-6 Lunawon 5 · lost 14

Gemini 3.1 Prowon 4 · lost 12

GPT-5won 5 · lost 16

DeepSeek V4 Prowon 5 · lost 16

GLM-5.2won 5 · lost 16

Claude Haiku 5.5won 3 · lost 10

Gemini 3 Flashwon 3 · lost 11

Claude Opus 4.5won 3 · lost 12

GPT-5.6 Lunawon 4 · lost 17

GLM-5.1won 4 · lost 17

Claude Opus 4.6won 3 · lost 14

MiMo V2.6 Prowon 3 · lost 15

O3won 3 · lost 17

GPT-5.2won 3 · lost 17

Grok 4.6won 3 · lost 18

Grok 4.5won 3 · lost 18

Gemini 3 Prowon 2 · lost 15

Grok 4.7won 2 · lost 16

Claude Sonnet 4.6won 2 · lost 17

Claude Sonnet 5won 2 · lost 19

GPT-5.4won 2 · lost 19

Qwen3.8 Maxwon 2 · lost 19

GPT-5.6 Solwon 2 · lost 19

Muse Spark 1.1won 2 · lost 19

Gemini 3.8 Flashwon 2 · lost 19

Claude Opus 4.7won 2 · lost 19

Gemini 4 Argonwon 1 · lost 12

Claude Opus 5.5won 1 · lost 18

GPT-5.6 Terrawon 1 · lost 20

GPT-5.5won 1 · lost 20

Claude Opus 4.8won 1 · lost 20

Claude Opus 5won 1 · lost 20

Claude Sonnet 5.5won 0 · lost 13

GPT-6.1 Solwon 0 · lost 13

GPT-6 Astrawon 0 · lost 14

DeepSeek V4.1 Flashwon 0 · lost 14

Muse Spark 1.2won 0 · lost 14

GLM-5.3won 0 · lost 15

Kimi K3won 0 · lost 15

Gemini 3.7 Flashwon 0 · lost 15

Claude Fable 5.1won 0 · lost 15

Claude Fable 5won 0 · lost 15

§ 3 · Sources

Where the numbers come from

5 publications, 22 figures. Every one links to the page it was read from.

LMArena Text ↗1 measureread 2026-11-10

LMArena Text 1467

GDPval-AA ↗1 measureread 2026-11-10

GDPval-AA 1129

AA-Briefcase ↗1 measureread 2026-11-10

AA-Briefcase 884

Vals.ai ↗13 measuresread 2026-11-10

Vals · Legal Research Bench 15.87%Vals · LegalBench 77.13%Vals · Finance Agent 41.5%Vals · CorpFin 61.42%Vals · TaxEval 73.79%Vals · MedCode 32.48%Vals · MedScribe 83.73%Vals · SWE-bench Verified 74% (Mini-SWE-agent)Vals · Vibe Code Bench 34.11% (OpenHands)Vals · Code Migration 21.56%Vals · GPQA Diamond 82.58%Vals · MMLU Pro 84.59%Vals · ProofBench 22%

Enkrypt AI Safety Leaderboard ↗6 measuresread 2026-11-10

Enkrypt · Jailbreak risk 24.5%Enkrypt · Harmful content risk 2.2%Enkrypt · CBRN risk 47%Enkrypt · Toxicity risk 1.8%Enkrypt · Bias risk 71.1%Enkrypt · Insecure code risk 17.3%

Badge

[![PublicAI Index](https://publicai.io/model-index/badge?model=mimo-v2-5-pro)](https://publicai.io/model-index/m/mimo-v2-5-pro)