Recent studyBlind TTS Elo is live. Compare two anonymous voice samples, vote after listening, and help separate real preference signal from noise.Vote in the study ->
Codesota · Models · MiniCPM-o 4.5-Instruct11 results · 11 benchmarks
Model card

MiniCPM-o 4.5-Instruct.

unknown
§ 02 · Benchmarks

Every benchmark MiniCPM-o 4.5-Instruct has a recorded score for.

#BenchmarkArea · TaskMetricValueRankDateSource
01MMAUMultimodal · Audio-Text-to-Textaccuracy76.9%#2/4source ↗
02HumanEval+Computer Code · Code Generationpass-186.6%#4/12source ↗
03MBPP+Computer Code · Code Generationpass-176.7%#5/9source ↗
04BIG-Bench HardReasoning · Multi-step Reasoningaccuracy81.1%#8/11source ↗
05TextVQAMultimodal · Visual Question Answeringaccuracy83.8%#8/23source ↗
06DocVQAComputer Vision · Document Understandinganls94.7%#9/21source ↗
07MMStarMultimodal · Image-Text-to-Textaccuracy73.1%#11/21source ↗
08MMMUMultimodal · Image-Text-to-Textaccuracy67.6%#18/36source ↗
09Video-MMEMultimodal · Video Understandingaccuracy70.4%#19/24source ↗
10GSM8KReasoning · Mathematical Reasoningaccuracy94.5%#27/48source ↗
11MMLUReasoning · Commonsense Reasoningaccuracy77.0%#52/64source ↗
Rank column shows this model’s position vs all other models scored on the same benchmark + metric (competitors after the slash). #1 in red means current SOTA. Sorted by rank, then newest result.
§ 03 · Strengths by area

Where MiniCPM-o 4.5-Instruct actually performs.

Computer Code
2
benchmarks
avg rank #4.5
Computer Vision
1
benchmark
avg rank #9.0
Multimodal
5
benchmarks
avg rank #11.6
Reasoning
3
benchmarks
avg rank #29.0
§ 04 · Papers

1 paper with results for MiniCPM-o 4.5-Instruct.

  1. 2026-04-30· 11 results

    MiniCPM-o 4.5: Towards Real-Time Full-Duplex Omni-Modal Interaction

§ 06 · Sources & freshness

Where these numbers come from.

pwc-dump
11
results
0 of 11 rows marked verified.