llmboard.aiAI model intelligence
Home

Model Rankings

OverallOpen ModelsAgentCodingReasoningMathKnowledgeInstruction FollowingTextVision
Image GenerationImage Editing
Video GenerationImage to VideoVideo Editing
Text to SpeechSpeech to Text
Embeddings

Efficiency

Chat Token PricingImage PricingVideo PricingAudio Pricing
Chat Speed & LatencyProvider Reliability

Benchmarks

GPQAMMLU-ProAIME 2025SWE-Bench VerifiedMMLUHumanity's Last ExamLiveCodeBenchMATHHumanEvalMMMU-Pro
All Benchmarks

Tools

Model Directory

Scoring & Data

Scoring & Data
1224 models729 benchmarks

Leaderboard Center

Overall RankingCodingCore BenchmarksPrice & ValueRuntime Performance

Modalities

All ModelsImage GenerationImage EditingVideo GenerationImage-to-VideoVideo EditingText-to-SpeechSpeech-to-TextEmbeddings

Data & Methods

Scoring MethodAll Benchmarks
llmboard.aiCopyright 2026 llmboard.ai

math benchmark

Humanity's Last Exam Leaderboard

Humanity's Last Exam (HLE) is a multi-modal academic benchmark with 2,500 questions across mathematics, humanities, and natural sciences.

Updated Sep 5, 2026

Models100
Model coverage103
MetricScore
EvidenceB

On this page

  • Ranking
  • Highlights
  • Distribution
  • Top models
  • About
  • FAQ

Humanity's Last Exam Ranking

Higher score ranks better on this benchmark.

30 of 100 rows
Columns

Show columns

Sort by
Rank
Model
Score
Percentile
Participants
Evidence
Evaluated
Rank01ModelANClaude Fable 5.1AnthropicScore65.00%Percentile100.00%Participants103EvidenceCEvaluatedSep 8, 2026
Rank02ModelANClaude Opus 5AnthropicScore64.70%Percentile99.02%Participants103EvidenceCEvaluatedSep 8, 2026
Rank03ModelANClaude Mythos PreviewAnthropicScore64.70%Percentile98.04%Participants103EvidenceCEvaluatedSep 8, 2026
Rank04ModelANClaude Fable 5AnthropicScore64.50%Percentile97.06%Participants103EvidenceCEvaluatedSep 8, 2026
Rank05ModelZAGLM-5.3Zhipu AIScore62.50%Percentile96.08%Participants103EvidenceCEvaluatedSep 8, 2026
Rank06ModelMEMuse Spark 1.1MetaScore62.10%Percentile95.10%Participants103EvidenceCEvaluatedSep 8, 2026
Rank07ModelDEDeepSeek-V4-Pro-0813DeepSeekScore60.00%Percentile94.12%Participants103EvidenceCEvaluatedSep 8, 2026
Rank08ModelMEMuse SparkMetaScore58.40%Percentile93.14%Participants103EvidenceCEvaluatedSep 8, 2026
Rank09ModelANClaude Opus 4.8AnthropicScore57.90%Percentile92.16%Participants103EvidenceCEvaluatedSep 8, 2026
Rank10ModelANClaude Sonnet 5AnthropicScore57.40%Percentile91.18%Participants103EvidenceCEvaluatedSep 8, 2026
Rank11ModelOPGPT-5.5 ProOpenAIScore57.20%Percentile90.20%Participants103EvidenceCEvaluatedSep 8, 2026
Rank12ModelMAKimi K3Moonshot AIScore56.00%Percentile89.22%Participants103EvidenceCEvaluatedSep 8, 2026
Rank13ModelBYSeed 2.1 ProByteDanceScore55.70%Percentile88.24%Participants103EvidenceCEvaluatedSep 8, 2026
Rank14ModelZAGLM-5.3-FlashZhipu AIScore55.30%Percentile87.25%Participants103EvidenceCEvaluatedSep 8, 2026
Rank15ModelANClaude Opus 4.7AnthropicScore54.70%Percentile86.27%Participants103EvidenceCEvaluatedSep 8, 2026
Rank16ModelZAGLM-5.2Zhipu AIScore54.70%Percentile85.29%Participants103EvidenceCEvaluatedSep 8, 2026
Rank17ModelBYSeed 2.1 TurboByteDanceScore54.60%Percentile84.31%Participants103EvidenceCEvaluatedSep 8, 2026
Rank18ModelANClaude Opus 4.6AnthropicScore53.10%Percentile83.33%Participants103EvidenceCEvaluatedSep 8, 2026
Rank19ModelZAGLM-5.1Zhipu AIScore52.30%Percentile82.35%Participants103EvidenceCEvaluatedSep 8, 2026
Rank20ModelOPGPT-5.5OpenAIScore52.20%Percentile81.37%Participants103EvidenceCEvaluatedSep 8, 2026
Rank21ModelGOGemini 3.1 ProGoogleScore51.40%Percentile80.39%Participants103EvidenceCEvaluatedSep 8, 2026
Rank22ModelMAKimi K2-Thinking-0905Moonshot AIScore51.00%Percentile79.41%Participants103EvidenceCEvaluatedSep 8, 2026
Rank23ModelXAGrok-4 HeavyxAIScore50.70%Percentile78.43%Participants103EvidenceCEvaluatedSep 8, 2026
Rank24ModelMAKimi K2.5Moonshot AIScore50.20%Percentile77.45%Participants103EvidenceCEvaluatedSep 8, 2026
Rank25ModelANClaude Sonnet 4.6AnthropicScore49.00%Percentile76.47%Participants103EvidenceCEvaluatedSep 8, 2026
Rank26ModelACQwen3.5-27BAlibaba Cloud / Qwen TeamScore48.50%Percentile75.49%Participants103EvidenceCEvaluatedSep 8, 2026
Rank27ModelDEDeepSeek-V4-Pro-MaxDeepSeekScore48.20%Percentile74.51%Participants103EvidenceCEvaluatedSep 8, 2026
Rank28ModelACQwen3.5-122B-A10BAlibaba Cloud / Qwen TeamScore47.50%Percentile73.53%Participants103EvidenceCEvaluatedSep 8, 2026
Rank29ModelACQwen3.5-35B-A3BAlibaba Cloud / Qwen TeamScore47.40%Percentile72.55%Participants103EvidenceCEvaluatedSep 8, 2026
Rank30ModelGOGemini 3 ProGoogleScore45.80%Percentile71.57%Participants103EvidenceCEvaluatedSep 8, 2026

Humanity's Last Exam Highlights

The leading models and scores on this benchmark.

Humanity's Last Exam Score Distribution

A closer view of the leading scores on this benchmark.

Humanity's Last Exam

The Top AI Models for Humanity's Last Exam

The first five results on this benchmark, with official price and output speed added where the model identity can be matched.

What is Humanity's Last Exam?

What Humanity's Last Exam measures and how its scores work.

Humanity's Last Exam (HLE) is a multi-modal academic benchmark covering mathematics, humanities, and natural sciences.

It measures LLM capabilities at the frontier of human knowledge using questions with unambiguous, verifiable solutions, reported as Score in ratio units.

Scores are shown in ratio. This benchmark is not independently verified and has an evidence level of B.

Family
Humanity's Last Exam
Modality
multimodal
Primary category
math
Score direction
higher
LLMBoard eligible
No
Evaluation key
overall

LLM Stats. Benchmark scores retain their original unit. Overall score eligibility is shown separately.

FAQ

Common questions about Humanity's Last Exam.

Which model scores highest on Humanity's Last Exam?

Claude Fable 5.1 is currently ranked first with 65.00%.

What are the top three models on Humanity's Last Exam?

The current leaders are Claude Fable 5.1 (65.00%), Claude Opus 5 (64.70%), and Claude Mythos Preview (64.70%).

Which Humanity's Last Exam model has the lowest official input price?

GPT-5 nano has the lowest matched official input price at $0.05 input / $0.40 output per 1M tokens.

Which models are fastest among Humanity's Last Exam results?

The fastest matched records are GPT OSS 20B (1,000.00 tok/s via Groq), GPT OSS 120B (500.00 tok/s via Groq), and GPT-5 nano (500.00 tok/s via OpenAI).

Does the highest score result prove overall model quality?

No. This benchmark measures one defined capability or task. The overall LLMBoard score uses a separate aggregation across eligible benchmark evidence.

What does Humanity's Last Exam measure?

It measures LLM capabilities at the frontier of human knowledge using questions with unambiguous, verifiable solutions, reported as Score in ratio units.

Is a higher score better?

Yes. Higher values rank better for this benchmark.

How many models are compared?

100 model results are currently shown.

Does this benchmark affect the overall score?

No. This benchmark is shown for reference but does not contribute to the overall score.

Rank #1Claude Fable 5.165.00%
Rank #2Claude Opus 564.70%
Rank #3Claude Mythos Preview64.70%
Rank #4Claude Fable 564.50%

Ranking basisThis humanity's last exam AI model leaderboard uses descending score in the benchmark original unit. The leaderboard ranking keeps matched price and speed data separate from benchmark evidence.

  1. 01
    AN
    Claude Fable 5.1Anthropic
    Score
    65.00%
    Price
    $10 input / $50 output per 1M tokens
    Speed
    Up to 7.67 tok/s via Anthropic

    Strengths

    • Ranks #1 of 103 compared models
    • 100th percentile on this benchmark
    • C evidence result

    Considerations

    • This result measures Humanity's Last Exam, not total model capability
  2. 02
    AN
    Claude Opus 5Anthropic
    Score
    64.70%
    Price
    $5.0 input / $25 output per 1M tokens
    Speed
    Up to 62.88 tok/s via Anthropic

    Strengths

    • Ranks #2 of 103 compared models
    • 99th percentile on this benchmark
    • C evidence result

    Considerations

    • This result measures Humanity's Last Exam, not total model capability
  3. 03
    AN
    Anthropic
    Score
    64.70%

    Strengths

    • Ranks #3 of 103 compared models
    • 98th percentile on this benchmark
    • C evidence result

    Considerations

    • This result measures Humanity's Last Exam, not total model capability
  4. 04
    AN
    Anthropic
    Score
    64.50%
    Price
    $10 input / $50 output per 1M tokens
    Speed
    Up to 43.18 tok/s via Anthropic

    Strengths

    • Ranks #4 of 103 compared models
    • 97th percentile on this benchmark
    • C evidence result

    Considerations

    • This result measures Humanity's Last Exam, not total model capability
  5. 05
    ZA
    Zhipu AI
    Score
    62.50%
    Price
    $1.4 input / $4.4 output per 1M tokens
    Speed
    Up to 472.69 tok/s via FriendliAI

    Strengths

    • Ranks #5 of 103 compared models
    • 96th percentile on this benchmark
    • C evidence result

    Considerations

    • This result measures Humanity's Last Exam, not total model capability

Selection summary

Best AI Models for Humanity's Last Exam

Claude Fable 5.1 currently leads Humanity's Last Exam with 65.00%. 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.

Claude Mythos Preview
Claude Fable 5
GLM-5.3
Benchmark rank #1Claude Fable 5.165.00% · $10 input / $50 output per 1M tokens
Benchmark rank #2Claude Opus 564.70% · $5.0 input / $25 output per 1M tokens
Benchmark rank #3Claude Mythos Preview64.70%