CritPt ↗4%
MiMo-V2.5-Pro · CritPt · Accuracy · % ↗
Xiaomi / MODEL EVIDENCE
MiMo V2.5 Pro
24 benchmarks · 25 observations · 24 sources · released 2026-04-23
Scores at a glance
24 benchmarks · large dot: MiMo V2.5 Pro; small dots: closest models where reported
ProofBench ↗22 %
MiMo V2.5 Pro · ProofBench · Accuracy ↗
1 more reported result
22% · MiMo-V2.5-Pro · ProofBench · Accuracy · % · Epoch AI · ProofBench ↗
LMCA ↗29.5%
MiMo-V2.5-Pro · LMCA · Score · % ↗
DTBench ↗84.53%
MiMo-V2.5-Pro · DTBench · Accuracy · % ↗
SciCode ↗50.23%
MiMo-V2.5-Pro · SciCode · Score · % ↗
Text Arena (Coding) ↗1474.92 Elo
MiMo-V2.5-Pro · WebDev Arena · Elo ↗
Legal Research Bench ↗15.87 %
MiMo V2.5 Pro · Legal Research Bench · All-Pass ↗
LegalBench ↗77.13 %
MiMo V2.5 Pro · LegalBench · Accuracy ↗
MiMo V2.5 Pro · Harvey's Legal Agent Benchmark v1 · Accuracy ↗
Tax Agent Bench ↗11.5 %
MiMo V2.5 Pro · Tax Agent Bench · Accuracy ↗
Vals Finance Agent v2 ↗41.5 %
MiMo V2.5 Pro · Finance Agent v2 · Accuracy ↗
MiMo V2.5 Pro · Excel Modeling Benchmark v1 · Accuracy ↗
CorpFin v2 ↗61.42 %
MiMo V2.5 Pro · CorpFin · Accuracy ↗
MortgageTax & TaxEval v2 ↗73.79 %
MiMo V2.5 Pro · TaxEval · Accuracy ↗
MedCode ↗32.48 %
MiMo V2.5 Pro · MedCode · Accuracy ↗
MedScribe ↗83.73 %
MiMo V2.5 Pro · MedScribe · Accuracy ↗
GPQA Diamond ↗82.58 %
MiMo V2.5 Pro · GPQA Diamond · Accuracy ↗
MMLU-Pro ↗84.59 %
MiMo V2.5 Pro · MMLU Pro · Accuracy ↗
Vibe Code Bench ↗34.11 %
MiMo V2.5 Pro (OpenHands) · Vibe Code Bench · original task set ↗
Terminal-Bench 4.0 ↗0.51 %
MiMo V2.5 Pro (Mini-SWE-agent) · Terminal-Bench 4.0 · Score ↗
Code Migration v1.1 ↗21.56 %
MiMo V2.5 Pro · Code Migration v1 · Accuracy ↗
MiMo V2.5 Pro (Mini-SWE-agent) · SWE-bench Verified · Score ↗
LiveCodeBench ↗81.35 %
MiMo V2.5 Pro · LiveCodeBench · Accuracy ↗
ALE-Bench ↗899.8 Performance points
mimo-v2.5-pro · Average contest performance ↗
Closest models by score
Similar positions on shared benchmark leaderboards. Smaller gaps are closer.
MiMo V2.5 →Xiaomi
15 of 18 shared benchmark comparisons
GPT-5.4 Mini →OpenAI
11 of 18 shared benchmark comparisons
15 of 18 shared benchmark comparisons
How similarity is measured
Mean absolute gap in percentile positions on shared benchmark settings; lower is closer. At least 60% overlap is required, with a minimum of three when available. Independent evaluations are used when there are at least three; otherwise published report comparisons are used. Repeated model observations within one setting are averaged before ranking. Writing-style traits and composite indices are excluded. This describes similarity, not an overall quality ranking.
Sources
Every plotted observation links to its evaluation. Reported configurations, benchmark versions and task subsets can differ; similarity describes shared scores, not training lift.
- Arena leaderboard — WebDev evaluation ↗
- SciCode Benchmark | Artificial Analysis ↗
- DTBench — Conceptual Reasoning Index ↗
- LMCA — Conceptual Reasoning Index ↗
- Ale Bench — methodology and published results ↗
- CritPt v1 ↗
- Code Migration v1 — evaluation methodology and results ↗
- CorpFin — evaluation methodology and results ↗
- Excel Modeling Benchmark v1 — evaluation methodology and results ↗
- Finance Agent v2 — evaluation methodology and results ↗
- GPQA Diamond — evaluation methodology and results ↗
- Harvey's Legal Agent Benchmark v1 — evaluation methodology and results ↗
- LiveCodeBench — evaluation methodology and results ↗
- LegalBench — evaluation methodology and results ↗
- Legal Research Bench — evaluation methodology and results ↗
- MedCode — evaluation methodology and results ↗
- MedScribe — evaluation methodology and results ↗
- MMLU Pro — evaluation methodology and results ↗
- ProofBench — evaluation methodology and results ↗
- SWE-bench Verified — evaluation methodology and results ↗
- Tax Agent Bench — evaluation methodology and results ↗
- TaxEval — evaluation methodology and results ↗
- Terminal-Bench 4.0 — evaluation methodology and results ↗
- Vibe Code Bench — evaluation methodology and results ↗
Names used in source tables
MiMo V2.5 Pro · MiMo V2.5 Pro (Mini-SWE-agent) · MiMo V2.5 Pro (OpenHands) · MiMo-V2.5-Pro · mimo-v2.5-pro