math benchmark
LiveBench is a benchmark with monthly releases of new questions based on recently-released datasets, arXiv papers, news articles, and IMDb movie synopses, covering math, coding, reasoning, language, instruction following, and data analysis.
Updated Sep 5, 2026
Higher score ranks better on this benchmark.
Rank | Model | Score | Percentile | Participants | Evidence | Evaluated |
|---|
| Rank01 | ModelOP | Score84.60% | Percentile100.00% | Participants38 | EvidenceC | Evaluated |
| Rank02 | ModelOP | Score80.71% | Percentile97.30% | Participants38 | EvidenceB | Evaluated |
| Rank03 | ModelOP | Score80.28% | Percentile94.59% | Participants38 | EvidenceB | Evaluated |
| Rank04 | ModelGO | Score79.93% | Percentile91.89% | Participants38 | EvidenceB | Evaluated |
| Rank05 | ModelAN | Score78.31% | Percentile89.19% | Participants38 | EvidenceB | Evaluated |
| Rank06 | ModelAN | Score77.22% | Percentile86.49% | Participants38 | EvidenceB | Evaluated |
| Rank07 | ModelAC | Score77.10% | Percentile83.78% | Participants38 | EvidenceC | Evaluated |
| Rank08 | ModelAN | Score76.91% | Percentile81.08% | Participants38 | EvidenceB | Evaluated |
| Rank09 | ModelMA | Score76.40% | Percentile78.38% | Participants38 | EvidenceC | Evaluated |
| Rank10 | ModelMA | Score76.40% | Percentile75.68% | Participants38 | EvidenceC | Evaluated |
| Rank11 | ModelAN | Score76.33% | Percentile72.97% | Participants38 | EvidenceB | Evaluated |
| Rank12 | ModelAN | Score75.96% | Percentile70.27% | Participants38 | EvidenceB | Evaluated |
| Rank13 | ModelAN | Score75.47% | Percentile67.57% | Participants38 | EvidenceB | Evaluated |
| Rank14 | ModelGO | Score75.02% | Percentile64.86% | Participants38 | EvidenceB | Evaluated |
| Rank15 | ModelAC | Score74.90% | Percentile62.16% | Participants38 | EvidenceC | Evaluated |
| Rank16 | ModelOP | Score74.84% | Percentile59.46% | Participants38 | EvidenceB | Evaluated |
| Rank17 | ModelOP | Score74.30% | Percentile56.76% | Participants38 | EvidenceB | Evaluated |
| Rank18 | ModelAC | Score74.30% | Percentile54.05% | Participants38 | EvidenceC | Evaluated |
| Rank19 | ModelAC | Score74.29% | Percentile51.35% | Participants38 | EvidenceB | Evaluated |
| Rank20 | ModelDE | Score73.58% | Percentile48.65% | Participants38 | EvidenceB | Evaluated |
| Rank21 | ModelGO | Score73.39% | Percentile45.95% | Participants38 | EvidenceB | Evaluated |
| Rank22 | ModelAC | Score73.10% | Percentile43.24% | Participants38 | EvidenceC | Evaluated |
| Rank23 | ModelOP | Score72.76% | Percentile40.54% | Participants38 | EvidenceB | Evaluated |
| Rank24 | ModelGO | Score72.40% | Percentile37.84% | Participants38 | EvidenceB | Evaluated |
| Rank25 | ModelMA | Score72.17% | Percentile35.14% | Participants38 | EvidenceB | Evaluated |
| Rank26 | ModelOP | Score72.04% | Percentile32.43% | Participants38 | EvidenceB | Evaluated |
| Rank27 | ModelMA | Score71.89% | Percentile29.73% | Participants38 | EvidenceB | Evaluated |
| Rank28 | ModelAC | Score70.85% | Percentile27.03% | Participants38 | EvidenceB | Evaluated |
| Rank29 | ModelZA | Score70.18% | Percentile24.32% | Participants38 | EvidenceB | Evaluated |
| Rank30 | ModelOP | Score70.13% | Percentile21.62% | Participants38 | EvidenceB | Evaluated |
The leading models and scores on this benchmark.
A closer view of the leading scores on this benchmark.
The first five results on this benchmark, with official price and output speed added where the model identity can be matched.
What LiveBench measures and how its scores work.
LiveBench is a contamination-limited LLM benchmark comprising tasks across math, coding, reasoning, language, instruction following, and data analysis.
It measures performance on tasks with verifiable, objective ground-truth answers, reported as Score in ratio units.
Scores are shown in ratio. This benchmark is not independently verified and has an evidence level of B.
LLM Stats. Benchmark scores retain their original unit. Overall score eligibility is shown separately.
Common questions about LiveBench.
o3-mini is currently ranked first with 84.60%.
The current leaders are o3-mini (84.60%), GPT-5.5 (80.71%), and GPT-5.4 (80.28%).
Qwen2.5-Omni-7B has the lowest matched official input price at $0.10 input / $0.40 output per 1M tokens.
The fastest matched records are Gemini 3.5 Flash (210.94 tok/s via Google), GPT-5.5 (134.94 tok/s via OpenAI), and Gemini 3 Flash (124.32 tok/s via Google).
No. This benchmark measures one defined capability or task. The overall LLMBoard score uses a separate aggregation across eligible benchmark evidence.
It measures performance on tasks with verifiable, objective ground-truth answers, reported as Score in ratio units.
Yes. Higher values rank better for this benchmark.
38 model results are currently shown.
No. This benchmark is shown for reference but does not contribute to the overall score.
Ranking basisThis livebench AI model leaderboard uses descending score in the benchmark original unit. The leaderboard ranking keeps matched price and speed data separate from benchmark evidence.
Selection summary
o3-mini currently leads LiveBench with 84.60%. It is the top model on this specific benchmark, while the best LLM for the broader task should also be checked against other benchmarks, price, and runtime.
Use this leaderboard with the supporting benchmark results and coverage details above. A leaderboard position summarizes the selected ranking signal; it does not replace workload-specific testing.