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 DirectoryCompare Models

Scoring & Data

Scoring & Data
393 models668 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 BenchmarksReasoningMath

Vendors

All VendorsOpenAIAnthropicGoogle
llmboard.aiCopyright 2026 llmboard.ai

reasoning benchmark

BIG-Bench Extra Hard Leaderboard

BIG-Bench Extra Hard (BBEH) is a challenging benchmark that replaces each task in BIG-Bench Hard with a novel task that probes similar reasoning capabilities but exhibits significantly increased difficulty. The benchmark contains 23 tasks testing diverse reasoning skills including many-hop reasoning, causal understanding, spatial reasoning, temporal arithmetic, geometric reasoning, linguistic reasoning, logic puzzles, and humor understanding. Designed to address saturation on existing benchmarks where state-of-the-art models achieve near-perfect scores, BBEH shows substantial room for improvement with best models achieving only 9.8-44.8% average accuracy.

Updated Aug 17, 2026

Models11
Model coverage11
MetricScore
EvidenceB

On this page

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

BIG-Bench Extra Hard Ranking

Higher score ranks better on this benchmark.

11 rows
Columns

Show columns

Sort by
Rank
Model
Score
Percentile
Participants
Evidence
Evaluated
Rank01ModelGOGemma 4 31BGoogleScore74.4%Percentile100.0%Participants11EvidenceCEvaluatedAug 17, 2026
Rank02ModelGOGemma 4 26B-A4BGoogleScore64.8%Percentile90.0%Participants11EvidenceCEvaluatedAug 17, 2026
Rank03ModelGOGemma 4 12BGoogleScore53.0%Percentile80.0%Participants11EvidenceCEvaluatedAug 17, 2026
Rank04ModelGODiffusionGemma 26B-A4BGoogleScore47.6%Percentile70.0%Participants11EvidenceCEvaluatedAug 17, 2026
Rank05ModelGOGemma 4 E4BGoogleScore33.1%Percentile60.0%Participants11EvidenceCEvaluatedAug 17, 2026
Rank06ModelGOGemma 4 E2BGoogleScore21.9%Percentile50.0%Participants11EvidenceCEvaluatedAug 17, 2026
Rank07ModelGOGemma 3 27BGoogleScore19.3%Percentile40.0%Participants11EvidenceCEvaluatedAug 17, 2026
Rank08ModelGOGemma 3 12BGoogleScore16.3%Percentile30.0%Participants11EvidenceCEvaluatedAug 17, 2026
Rank09ModelGOGemini DiffusionGoogleScore15.0%Percentile20.0%Participants11EvidenceCEvaluatedAug 17, 2026
Rank10ModelGOGemma 3 4BGoogleScore11.0%Percentile10.0%Participants11EvidenceCEvaluatedAug 17, 2026
Rank11ModelGOGemma 3 1BGoogleScore7.2%Percentile0.0%Participants11EvidenceCEvaluatedAug 17, 2026

BIG-Bench Extra Hard Highlights

The leading models and scores on this benchmark.

Rank #1Gemma 4 31B74.4%Rank #2Gemma 4 26B-A4B64.8%Rank #3Gemma 4 12B53.0%Rank #4DiffusionGemma 26B-A4B47.6%

BIG-Bench Extra Hard Score Distribution

A closer view of the leading scores on this benchmark.

BIG-Bench Extra Hard

The Top AI Models for BIG-Bench Extra Hard

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

Ranking basisThis big-bench extra hard 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.

  1. 01
    GO
    Gemma 4 31BGoogle
    Score
    74.4%
    Speed
    Up to 8.3 tok/s via FriendliAI

    Strengths

    • Ranks #1 of 11 compared models
    • 100th percentile on this benchmark
    • C evidence result

    Considerations

    • This result measures BIG-Bench Extra Hard, not total model capability
  2. 02
    GO
    Gemma 4 26B-A4BGoogle
    Score
    64.8%

    Strengths

    • Ranks #2 of 11 compared models
    • 90th percentile on this benchmark
    • C evidence result

    Considerations

    • This result measures BIG-Bench Extra Hard, not total model capability
  3. 03
    GO
    Gemma 4 12BGoogle
    Score
    53.0%

    Strengths

    • Ranks #3 of 11 compared models
    • 80th percentile on this benchmark
    • C evidence result

    Considerations

    • This result measures BIG-Bench Extra Hard, not total model capability
  4. 04
    GO
    DiffusionGemma 26B-A4BGoogle
    Score
    47.6%

    Strengths

    • Ranks #4 of 11 compared models
    • 70th percentile on this benchmark
    • C evidence result

    Considerations

    • This result measures BIG-Bench Extra Hard, not total model capability
  5. 05
    GO
    Gemma 4 E4BGoogle
    Score
    33.1%

    Strengths

    • Ranks #5 of 11 compared models
    • 60th percentile on this benchmark
    • C evidence result

    Considerations

    • This result measures BIG-Bench Extra Hard, not total model capability

Selection summary

Best AI Models for BIG-Bench Extra Hard

Gemma 4 31B currently leads BIG-Bench Extra Hard with 74.4%. 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.

Benchmark rank #1Gemma 4 31B74.4% · Up to 8.3 tok/s via FriendliAIBenchmark rank #2Gemma 4 26B-A4B64.8%Benchmark rank #3Gemma 4 12B53.0%

What is BIG-Bench Extra Hard?

What BIG-Bench Extra Hard measures and how its scores work.

BIG-Bench Extra Hard (BBEH) is a challenging benchmark that replaces each task in BIG-Bench Hard with a novel task that probes similar reasoning capabilities but exhibits significantly increased difficulty. The benchmark contains 23 tasks testing diverse reasoning skills including many-hop reasoning, causal understanding, spatial reasoning, temporal arithmetic, geometric reasoning, linguistic reasoning, logic puzzles, and humor understanding. Designed to address saturation on existing benchmarks where state-of-the-art models achieve near-perfect scores, BBEH shows substantial room for improvement with best models achieving only 9.8-44.8% average accuracy.

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

Family
BIG-Bench Extra Hard
Modality
text
Primary category
reasoning
Score direction
higher
LLMBoard eligible
Yes
Evaluation key
big-bench-extra-hard|llm-stats-current

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

FAQ

Common questions about BIG-Bench Extra Hard.

Which model scores highest on BIG-Bench Extra Hard?

Gemma 4 31B is currently ranked first with 74.4%.

What does BIG-Bench Extra Hard measure?

BIG-Bench Extra Hard (BBEH) is a challenging benchmark that replaces each task in BIG-Bench Hard with a novel task that probes similar reasoning capabilities but exhibits significantly increased difficulty. The benchmark contains 23 tasks testing diverse reasoning skills including many-hop reasoning, causal understanding, spatial reasoning, temporal arithmetic, geometric reasoning, linguistic reasoning, logic puzzles, and humor understanding. Designed to address saturation on existing benchmarks where state-of-the-art models achieve near-perfect scores, BBEH shows substantial room for improvement with best models achieving only 9.8-44.8% average accuracy.

Is a higher score better?

Yes. Higher values rank better for this benchmark.

How many models are compared?

11 model results are currently shown.

Does this benchmark affect the overall score?

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