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
1367 models808 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

reasoning benchmark

AA GPQA Diamond Leaderboard

Diamond 198 questions; multiple choice; regex extraction; pass@1

Updated Sep 22, 2026

Models100
Model coverage199
Metricgpqa
EvidenceB

On this page

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

AA GPQA Diamond Ranking

Higher gpqa ranks better on this benchmark.

30 of 100 rows
Columns

Show columns

Sort by
Rank
Model
gpqa
Percentile
Participants
Evidence
Evaluated
Rank01ModelOPGPT-6 AstraOpenAIgpqa96.26%Percentile100.00%Participants199EvidenceBEvaluatedN/A
Rank02ModelGOGemini 3.8 FlashGooglegpqa95.25%Percentile99.49%Participants199EvidenceBEvaluatedN/A
Rank03ModelXAGrok 4.6xAIgpqa94.95%Percentile98.99%Participants199EvidenceBEvaluatedN/A
Rank04ModelGOGemini 3.7 FlashGooglegpqa94.55%Percentile98.48%Participants199EvidenceBEvaluatedN/A
Rank05ModelOPGPT-5.6 SolOpenAIgpqa94.14%Percentile97.98%Participants199EvidenceBEvaluatedN/A
Rank06ModelMEMuse Spark 1.3Metagpqa94.14%Percentile97.47%Participants199EvidenceBEvaluatedN/A
Rank07ModelGOGemini 3.1 ProGooglegpqa94.14%Percentile96.97%Participants199EvidenceBEvaluatedN/A
Rank08ModelANClaude Fable 5.1Anthropicgpqa93.74%Percentile96.46%Participants199EvidenceBEvaluatedN/A
Rank09ModelANClaude Opus 5Anthropicgpqa93.74%Percentile95.96%Participants199EvidenceBEvaluatedN/A
Rank10ModelOPGPT-5.5OpenAIgpqa93.54%Percentile95.45%Participants199EvidenceBEvaluatedN/A
Rank11ModelMAKimi K3Moonshot AIgpqa93.54%Percentile94.95%Participants199EvidenceBEvaluatedN/A
Rank12ModelXAGrok 4.5xAIgpqa93.13%Percentile94.44%Participants199EvidenceBEvaluatedN/A
Rank13ModelMIMiniMax M3MiniMaxgpqa92.93%Percentile93.94%Participants199EvidenceBEvaluatedN/A
Rank14ModelACQwen3.8 MaxAlibaba Cloud / Qwen Teamgpqa92.83%Percentile93.43%Participants199EvidenceBEvaluatedN/A
Rank15ModelGOGemini 3.6 FlashGooglegpqa92.83%Percentile92.93%Participants199EvidenceBEvaluatedN/A
Rank16ModelDEDeepSeek-V4-Pro-0813DeepSeekgpqa92.83%Percentile92.42%Participants199EvidenceBEvaluatedN/A
Rank17ModelANClaude Fable 5Anthropicgpqa92.63%Percentile91.92%Participants199EvidenceBEvaluatedN/A
Rank18ModelOPGPT-5.6 TerraOpenAIgpqa92.53%Percentile91.41%Participants199EvidenceBEvaluatedN/A
Rank19ModelACQwen3.7 MaxAlibaba Cloud / Qwen Teamgpqa92.32%Percentile90.91%Participants199EvidenceBEvaluatedN/A
Rank20ModelACQwen3.8-Flash-NextAlibaba Cloud / Qwen Teamgpqa92.32%Percentile90.40%Participants199EvidenceBEvaluatedN/A
Rank21ModelGOGemini 3.5 FlashGooglegpqa92.22%Percentile89.90%Participants199EvidenceBEvaluatedN/A
Rank22ModelOPGPT-5.4OpenAIgpqa92.02%Percentile89.39%Participants199EvidenceBEvaluatedN/A
Rank23ModelANClaude Opus 4.8Anthropicgpqa92.02%Percentile88.89%Participants199EvidenceBEvaluatedN/A
Rank24ModelZAGLM-5.3Zhipu AIgpqa91.72%Percentile88.38%Participants199EvidenceBEvaluatedN/A
Rank25ModelOPGPT-5.3 CodexOpenAIgpqa91.52%Percentile87.88%Participants199EvidenceBEvaluatedN/A
Rank26ModelANClaude Opus 4.7Anthropicgpqa91.41%Percentile87.37%Participants199EvidenceBEvaluatedN/A
Rank27ModelZAGLM-5.3-FlashZhipu AIgpqa91.21%Percentile86.87%Participants199EvidenceBEvaluatedN/A
Rank28ModelOPGPT-5.6 LunaOpenAIgpqa91.11%Percentile86.36%Participants199EvidenceBEvaluatedN/A
Rank29ModelMAKimi K2.6Moonshot AIgpqa91.11%Percentile85.86%Participants199EvidenceBEvaluatedN/A
Rank30ModelANClaude Sonnet 5Anthropicgpqa91.11%Percentile85.35%Participants199EvidenceBEvaluatedN/A

AA GPQA Diamond Highlights

The leading models and scores on this benchmark.

AA GPQA Diamond Score Distribution

A closer view of the leading scores on this benchmark.

AA GPQA Diamond

The Top AI Models for AA GPQA Diamond

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

What is AA GPQA Diamond?

What AA GPQA Diamond measures and how its scores work.

Diamond 198 questions; multiple choice; regex extraction; pass@1

Scores are shown in ratio. This benchmark is verified and has an evidence level of B.

Family
gpqa
Modality
text
Primary category
reasoning
Score direction
higher
LLMBoard eligible
Yes
Evaluation key
overall

Artificial Analysis. Benchmark scores retain their original unit. Overall score eligibility is shown separately.

FAQ

Common questions about AA GPQA Diamond.

Which model scores highest on AA GPQA Diamond?

GPT-6 Astra is currently ranked first with 96.26%.

What are the top three models on AA GPQA Diamond?

The current leaders are GPT-6 Astra (96.26%), Gemini 3.8 Flash (95.25%), and Grok 4.6 (94.95%).

Which AA GPQA Diamond model has the lowest official input price?

Step-3.5-Flash has the lowest matched official input price at $0.10 input / $0.30 output per 1M tokens.

Which models are fastest among AA GPQA Diamond results?

The fastest matched records are Gemini 3.7 Flash (409.96 tok/s via Google), GPT-5.6 Luna (215.52 tok/s via OpenAI), and GPT-5 mini (200.00 tok/s via OpenAI).

Does the highest gpqa 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 AA GPQA Diamond measure?

Diamond 198 questions; multiple choice; regex extraction; pass@1

Is a higher gpqa 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?

Yes. This benchmark can contribute to the current LLMBoard capability score.

Rank #1GPT-6 Astra96.26%
Rank #2Gemini 3.8 Flash95.25%
Rank #3Grok 4.694.95%
Rank #4Gemini 3.7 Flash94.55%

Ranking basisThis aa gpqa diamond AI model leaderboard uses descending gpqa in the benchmark original unit. The leaderboard ranking keeps matched price and speed data separate from benchmark evidence.

  1. 01
    OP
    GPT-6 AstraOpenAI
    gpqa
    96.26%
    Price
    $10 input / $50 output per 1M tokens

    Strengths

    • Ranks #1 of 199 compared models
    • 100th percentile on this benchmark
    • B evidence result

    Considerations

    • This result measures AA GPQA Diamond, not total model capability
  2. 02
    GO
    Gemini 3.8 FlashGoogle
    gpqa
    95.25%
    Price
    $0.75 input / $3.8 output per 1M tokens
    Speed
    Up to 55.66 tok/s via Google

    Strengths

    • Ranks #2 of 199 compared models
    • 99th percentile on this benchmark
    • B evidence result

    Considerations

    • This result measures AA GPQA Diamond, not total model capability
  3. 03
    XA
    xAI
    gpqa
    94.95%
    Price
    $2.0 input / $6.0 output per 1M tokens

    Strengths

    • Ranks #3 of 199 compared models
    • 99th percentile on this benchmark
    • B evidence result

    Considerations

    • This result measures AA GPQA Diamond, not total model capability
  4. 04
    GO
    Google
    gpqa
    94.55%
    Price
    $0.75 input / $3.8 output per 1M tokens
    Speed
    Up to 409.96 tok/s via Google

    Strengths

    • Ranks #4 of 199 compared models
    • 98th percentile on this benchmark
    • B evidence result

    Considerations

    • This result measures AA GPQA Diamond, not total model capability
  5. 05
    OP
    OpenAI
    gpqa
    94.14%
    Price
    $4.0 input / $20 output per 1M tokens
    Speed
    Up to 25.46 tok/s via OpenAI

    Strengths

    • Ranks #5 of 199 compared models
    • 98th percentile on this benchmark
    • B evidence result

    Considerations

    • This result measures AA GPQA Diamond, not total model capability

Selection summary

Best AI Models for AA GPQA Diamond

GPT-6 Astra currently leads AA GPQA Diamond with 96.26%. 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.

Grok 4.6
Gemini 3.7 Flash
GPT-5.6 Sol
Benchmark rank #1GPT-6 Astra96.26% · $10 input / $50 output per 1M tokens
Benchmark rank #2Gemini 3.8 Flash95.25% · $0.75 input / $3.8 output per 1M tokens
Benchmark rank #3Grok 4.694.95% · $2.0 input / $6.0 output per 1M tokens