GPQA Diamond ↗46.59%
GPT-4 Turbo (Apr 2024) · GPQA diamond · Best score (across scorers) · % ↗
OpenAI / MODEL EVIDENCE
GPT-4 Turbo (Apr 2024)
9 benchmarks · 12 observations · 11 sources · released 2024-04-09
Scores at a glance
9 benchmarks · large dot: GPT-4 Turbo (Apr 2024); small dots: closest models where reported
GPT-4 Turbo (Apr 2024) · FrontierMath-Tiers-1-3-v2-Private · Best score (across scorers) · % ↗
MATH Level 5 ↗46.73%
GPT-4 Turbo (Apr 2024) · MATH level 5 · Best score (across scorers) · % ↗
GPT-4 Turbo (Apr 2024) · OTIS Mock AIME 2024-2025 · Best score (across scorers) · % ↗
MMLU ↗81.3%
GPT-4 Turbo (Apr 2024) · MMLU · EM · % ↗
SimpleBench ↗25.1%
GPT-4 Turbo (Apr 2024) · SimpleBench · Score (AVG@5) · % ↗
WeirdML (v2) ↗18.01%
GPT-4 Turbo (Apr 2024) · WeirdML · Accuracy · % ↗
GPT-4 Turbo (Apr 2024) · METR task performance & time horizons · Average task accuracy ↗
3 more reported results
27.18% · GPT-4 Turbo (Apr 2024) · METR task performance & time horizons · Average task accuracy · Epoch AI · METR Time Horizons ↗
3.73279 minutes · GPT-4 Turbo (Apr 2024) · METR task performance & time horizons · 50% success task horizon · METR-Horizon-v1.1 · Epoch AI · METR Autonomy / Task-Horizon Suite · public result data ↗
6.56734 minutes · GPT-4 Turbo (Apr 2024) · METR task performance & time horizons · 50% success task horizon · Epoch AI · METR Autonomy / Task-Horizon Suite · public result data ↗
Epoch Capabilities Index (ECI) ↗127.27 ECI
GPT-4 Turbo (Apr 2024) · General ECI · 26 October 2026 fit ↗
Closest models by score
Similar positions on shared benchmark leaderboards. Smaller gaps are closer.
Mistral Small 3 →Mistral
3 of 4 shared benchmark comparisons
GPT-4o (Nov 2024) →OpenAI
3 of 4 shared benchmark comparisons
Mistral Small 3.1 →Mistral
3 of 4 shared benchmark comparisons
How similarity is measured
Mean absolute gap in percentile positions on shared benchmark settings; lower is closer. At least 60% overlap is required, with a minimum of three when available. Independent evaluations are used when there are at least three; otherwise published report comparisons are used. Repeated model observations within one setting are averaged before ranking. Writing-style traits and composite indices are excluded. This describes similarity, not an overall quality ranking.
Sources
Every plotted observation links to its evaluation. Reported configurations, benchmark versions and task subsets can differ; similarity describes shared scores, not training lift.
- Holistic Evaluation of Language Models (HELM) ↗
- FrontierMath Tiers 1-3 (v2) ↗
- GPQA Diamond ↗
- MATH Level 5 ↗
- METR Time Horizons ↗
- OTIS Mock AIME 2024-2025 ↗
- SimpleBench ↗
- AI benchmarks & capabilities ↗
- WeirdML ↗
- Measuring AI Ability to Complete Long Software Tasks ↗
- Metr Autonomy Task Horizon Suite — methodology and published results ↗
Names used in source tables
GPT-4 Turbo (Apr 2024)