math benchmark
LiveBench Leaderboard
LiveBench is a challenging, contamination-limited LLM benchmark that addresses test set contamination by releasing new questions monthly based on recently-released datasets, arXiv papers, news articles, and IMDb movie synopses. It comprises tasks across math, coding, reasoning, language, instruction following, and data analysis with verifiable, objective ground-truth answers.
Updated Aug 17, 2026
Models38
Model coverage38
MetricScore
EvidenceB
LiveBench Ranking
Higher score ranks better on this benchmark.
Sort by
Rank | Model | Score | Percentile | Participants | Evidence | Evaluated |
|---|
| Rank01 | ModelOP | Score84.6% | Percentile100.0% | Participants38 | EvidenceC | Evaluated |
| Rank02 | ModelOP | Score80.7% | Percentile97.3% | Participants38 | EvidenceB | Evaluated |
| Rank03 | ModelOP | Score80.3% | Percentile94.6% | Participants38 | EvidenceB | Evaluated |
| Rank04 | ModelGO | Score79.9% | Percentile91.9% | Participants38 | EvidenceB | Evaluated |
| Rank05 | ModelAN | Score78.3% | Percentile89.2% | Participants38 | EvidenceB | Evaluated |
| Rank06 | ModelAN | Score77.2% | Percentile86.5% | Participants38 | EvidenceB | Evaluated |
| Rank07 | ModelAC | Score77.1% | Percentile83.8% | Participants38 | EvidenceC | Evaluated |
| Rank08 | ModelAN | Score76.9% | Percentile81.1% | Participants38 | EvidenceB | Evaluated |
| Rank09 | ModelMA | Score76.4% | Percentile78.4% | Participants38 | EvidenceC | Evaluated |
| Rank10 | ModelMA | Score76.4% | Percentile75.7% | Participants38 | EvidenceC | Evaluated |
| Rank11 | ModelAN | Score76.3% | Percentile73.0% | Participants38 | EvidenceB | Evaluated |
| Rank12 | ModelAN | Score76.0% | Percentile70.3% | Participants38 | EvidenceB | Evaluated |
| Rank13 | ModelAN | Score75.5% | Percentile67.6% | Participants38 | EvidenceB | Evaluated |
| Rank14 | ModelGO | Score75.0% | Percentile64.9% | Participants38 | EvidenceB | Evaluated |
| Rank15 | ModelAC | Score74.9% | Percentile62.2% | Participants38 | EvidenceC | Evaluated |
| Rank16 | ModelOP | Score74.8% | Percentile59.5% | Participants38 | EvidenceB | Evaluated |
| Rank17 | ModelOP | Score74.3% | Percentile56.8% | Participants38 | EvidenceB | Evaluated |
| Rank18 | ModelAC | Score74.3% | Percentile54.0% | Participants38 | EvidenceC | Evaluated |
| Rank19 | ModelAC | Score74.3% | Percentile51.4% | Participants38 | EvidenceB | Evaluated |
| Rank20 | ModelDE | Score73.6% | Percentile48.6% | Participants38 | EvidenceB | Evaluated |
| Rank21 | ModelGO | Score73.4% | Percentile46.0% | Participants38 | EvidenceB | Evaluated |
| Rank22 | ModelAC | Score73.1% | Percentile43.2% | Participants38 | EvidenceC | Evaluated |
| Rank23 | ModelOP | Score72.8% | Percentile40.5% | Participants38 | EvidenceB | Evaluated |
| Rank24 | ModelGO | Score72.4% | Percentile37.8% | Participants38 | EvidenceB | Evaluated |
| Rank25 | ModelMA | Score72.2% | Percentile35.1% | Participants38 | EvidenceB | Evaluated |
| Rank26 | ModelOP | Score72.0% | Percentile32.4% | Participants38 | EvidenceB | Evaluated |
| Rank27 | ModelMA | Score71.9% | Percentile29.7% | Participants38 | EvidenceB | Evaluated |
| Rank28 | ModelAC | Score70.9% | Percentile27.0% | Participants38 | EvidenceB | Evaluated |
| Rank29 | ModelZA | Score70.2% | Percentile24.3% | Participants38 | EvidenceB | Evaluated |
| Rank30 | ModelOP | Score70.1% | Percentile21.6% | Participants38 | EvidenceB | Evaluated |
| Rank31 | ModelMI | Score70.0% | Percentile18.9% | Participants38 | EvidenceB | Evaluated |
| Rank32 | ModelMA | Score69.1% | Percentile16.2% | Participants38 | EvidenceB | Evaluated |
| Rank33 | ModelOP | Score67.0% | Percentile13.5% | Participants38 | EvidenceC | Evaluated |
| Rank34 | ModelOP | Score52.3% | Percentile10.8% | Participants38 | EvidenceC | Evaluated |
| Rank35 | ModelAC | Score52.3% | Percentile8.1% | Participants38 | EvidenceC | Evaluated |
| Rank36 | ModelMI | Score47.6% | Percentile5.4% | Participants38 | EvidenceC | Evaluated |
| Rank37 | ModelAC | Score35.9% | Percentile2.7% | Participants38 | EvidenceC | Evaluated |
| Rank38 | ModelAC | Score29.6% | Percentile0.0% | Participants38 | EvidenceC | Evaluated |
LiveBench Highlights
The leading models and scores on this benchmark.
LiveBench Score Distribution
A closer view of the leading scores on this benchmark.
The Top AI Models for LiveBench
The first five results on this benchmark, with official price and output speed added where the model identity can be matched.
Ranking basisThis livebench AI model leaderboard uses descending score in the benchmark's original unit. The leaderboard ranking keeps matched price and speed data separate from benchmark evidence.
- 01OPo3-miniOpenAI
- Score
- 84.6%
- Price
- $1.1 input / $4.4 output per 1M tokens
- Speed
- Up to 115 tok/s via Azure
Strengths
- Ranks #1 of 38 compared models
- 100th percentile on this benchmark
- C evidence result
Considerations
- This result measures LiveBench, not total model capability
- 02OPGPT-5.5OpenAI
- Score
- 80.7%
- Price
- $5.0 input / $30 output per 1M tokens
Strengths
- Ranks #2 of 38 compared models
- 97th percentile on this benchmark
- B evidence result
Considerations
- This result measures LiveBench, not total model capability
- 03OPGPT-5.4OpenAI
- Score
- 80.3%
- Price
- $2.5 input / $15 output per 1M tokens
- Speed
- Up to 9.1 tok/s via OpenAI
Strengths
- Ranks #3 of 38 compared models
- 95th percentile on this benchmark
- B evidence result
Considerations
- This result measures LiveBench, not total model capability
- 04GOGemini 3.1 ProGoogle
- Score
- 79.9%
- Price
- $2.0 input / $12 output per 1M tokens
- Speed
- Up to 23 tok/s via Google
Strengths
- Ranks #4 of 38 compared models
- 92th percentile on this benchmark
- B evidence result
Considerations
- This result measures LiveBench, not total model capability
- 05ANClaude Fable 5Anthropic
- Score
- 78.3%
- Price
- $10 input / $50 output per 1M tokens
- Speed
- Up to 9.1 tok/s via Anthropic
Strengths
- Ranks #5 of 38 compared models
- 89th percentile on this benchmark
- B evidence result
Considerations
- This result measures LiveBench, not total model capability
Selection summary
Best AI Models for LiveBench
o3-mini currently leads LiveBench with 84.6%. 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.
What is LiveBench?
What LiveBench measures and how its scores work.
LiveBench is a challenging, contamination-limited LLM benchmark that addresses test set contamination by releasing new questions monthly based on recently-released datasets, arXiv papers, news articles, and IMDb movie synopses. It comprises tasks across math, coding, reasoning, language, instruction following, and data analysis with verifiable, objective ground-truth answers.
Scores are shown in ratio. This benchmark is not independently verified and has an evidence level of B.
- Family
- LiveBench
- Modality
- text
- Primary category
- math
- Score direction
- higher
- LLMBoard eligible
- Yes
- Evaluation key
- livebench|llm-stats-current
Benchmark scores retain their original unit. Overall score eligibility is shown separately.
FAQ
Common questions about LiveBench.
Which model scores highest on LiveBench?
o3-mini is currently ranked first with 84.6%.
What does LiveBench measure?
LiveBench is a challenging, contamination-limited LLM benchmark that addresses test set contamination by releasing new questions monthly based on recently-released datasets, arXiv papers, news articles, and IMDb movie synopses. It comprises tasks across math, coding, reasoning, language, instruction following, and data analysis with verifiable, objective ground-truth answers.
Is a higher score better?
Yes. Higher values rank better for this benchmark.
How many models are compared?
38 model results are currently shown.
Does this benchmark affect the overall score?
Yes. This benchmark can contribute to the current LLMBoard capability score.