RLTasksby ulam.ai
Models / OpenAI

OpenAI / MODEL EVIDENCE

GPT-6 Sol

58 benchmarks · 78 observations · 58 sources · released 2026-09-22

Scores at a glance

58 benchmarks · large dot: GPT-6 Sol; small dots: closest models where reported

AA-3D ↗1120.99 Elo
572.551198.83

GPT-6 Sol (Max) · AA-3D · Elo ↗

MMMU-Pro ↗82.95%
73.4787.69

GPT-6 Sol (Max) · MMMU-Pro · AA implementation ↗

HealthBench ↗62.34%
50.765.98

GPT-6 Sol (Max) · HealthBench · HealthBench Professional · length-adjusted ↗

GPQA Diamond ↗94.32%
9.2896.4

GPT-6 Sol (max) · GPQA diamond · Best score (across scorers) · % ↗

LMCA ↗59.08%
2.7868.26

GPT-6 Sol (max) · LMCA · Score · % ↗

DTBench ↗97.33%
41.6598.93

GPT-6 Sol (max) · DTBench · Accuracy · % ↗

GDP.pdf ↗26.4%
235

GPT-6 Sol (max) · GDP.pdf · GDP.pdf score · % ↗

SpireClimb ↗7.5%
025.67

GPT-6 Sol (max) · SpireClimb · Best score (across scorers) · % ↗

Vals RSI Index ↗0.42 reference score
00.46

GPT-6 Sol (max) · Compression · reference-anchored score ↗

3 more reported results

0.31 reference score · GPT-6 Sol (max) · LM training · reference-anchored score · Vals RSI Index · LM training ↗

0.23 reference score · GPT-6 Sol (max) · Harness engineering · reference-anchored score · Vals RSI Index · Harness engineering ↗

0.16 reference score · GPT-6 Sol (max) · Post-training · reference-anchored score · Vals RSI Index · Post-training ↗

Fortress ↗13.46%
5.7874.39

gpt-6-sol · Scale Labs · Risk elicitation score ↗

Closest models by score

Similar positions on shared benchmark leaderboards. Smaller gaps are closer.

13.4 percentile-point gap

31 of 44 shared benchmark comparisons

16.6 percentile-point gap

36 of 44 shared benchmark comparisons

19.0 percentile-point gap

36 of 44 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.

Filter comparisons to this model →

Sources

Every plotted observation links to its evaluation. Reported configurations, benchmark versions and task subsets can differ; similarity describes shared scores, not training lift.

  1. Blueprint-Bench 2 | Andon Labs ↗

    Andon Labs

  2. ARC Prize - Leaderboard ↗

    ARC Prize Foundation

  3. Arena leaderboard — WebDev evaluation ↗

    Arena

  4. AA-3D: Interactive 3D Experience Benchmark | Artificial Analysis ↗

    Artificial Analysis

  5. Harvey LAB-AA v1.1 Benchmark | Artificial Analysis ↗

    Artificial Analysis

  6. HealthBench Professional Benchmark | Artificial Analysis ↗

    Artificial Analysis

  7. ITBench-AA Benchmark | Artificial Analysis ↗

    Artificial Analysis

  8. MLCR-AA: Medical Long Context Reasoning Benchmark | Artificial Analysis ↗

    Artificial Analysis

  9. MMMU-Pro Benchmark | Artificial Analysis ↗

    Artificial Analysis

  10. SciCode Benchmark | Artificial Analysis ↗

    Artificial Analysis

  11. FrontierCode Leaderboard ↗

    Cognition

  12. DTBench — Conceptual Reasoning Index ↗

    Conceptual Reasoning Index

  13. LMCA — Conceptual Reasoning Index ↗

    Conceptual Reasoning Index

  14. APEX-Agents ↗

    Epoch AI

  15. ARC-AGI-2 ↗

    Epoch AI

  16. Chess Puzzles ↗

    Epoch AI

  17. CritPt v1 ↗

    Epoch AI

  18. Earthborne Rangers (EBR-bench) ↗

    Epoch AI

  19. FrontierMath Erdős ↗

    Epoch AI

  20. FrontierMath Tier 4 (v2) ↗

    Epoch AI

  21. FrontierMath Tiers 1-3 (v2) ↗

    Epoch AI

  22. Furniture Assembly Benchmark ↗

    Epoch AI

  23. GPQA Diamond ↗

    Epoch AI

  24. Mystery Game Puzzles ↗

    Epoch AI

  25. OTIS Mock AIME 2024-2025 ↗

    Epoch AI

  26. SimpleQA Verified ↗

    Epoch AI

  27. SpireClimb ↗

    Epoch AI

  28. Vending Bench 2 — methodology and published results ↗

    Epoch AI · 2026

  29. AI benchmarks & capabilities ↗

    Epoch AI

  30. WeirdML ↗

    Håvard Tveit Ihle

  31. DistressBench — evaluation methodology and results ↗

    Scale Labs / SEAL · 2026

  32. Scale Labs — evaluation methodology and results ↗

    Scale Labs / SEAL · 2026

  33. Scale Labs — evaluation methodology and results ↗

    Scale Labs / SEAL · 2026

  34. Humanity's Sixth Sense — evaluation methodology and results ↗

    Scale Labs / SEAL · 2026

  35. Scale Labs — evaluation methodology and results ↗

    Scale Labs / SEAL · 2026

  36. GDP.pdf Benchmark | Surge AI ↗

    Surge AI

  37. BioMysteryBench — evaluation methodology and results ↗

    Vals AI · 2026

  38. Code Migration v1 — evaluation methodology and results ↗

    Vals AI · 2026

  39. CyberBench — evaluation methodology and results ↗

    Vals AI · 2026

  40. Excel Modeling Benchmark v1 — evaluation methodology and results ↗

    Vals AI · 2026

  41. Finance Agent v2 — evaluation methodology and results ↗

    Vals AI · 2026

  42. Harvey's Legal Agent Benchmark v1 — evaluation methodology and results ↗

    Vals AI · 2026

  43. IOI — evaluation methodology and results ↗

    Vals AI · 2026

  44. Legal Research Bench — evaluation methodology and results ↗

    Vals AI · 2026

  45. MedCode — evaluation methodology and results ↗

    Vals AI · 2026

  46. MedScribe — evaluation methodology and results ↗

    Vals AI · 2026

  47. MysteryMechanism — evaluation methodology and results ↗

    Vals AI · 2026

  48. ProgramBench — evaluation methodology and results ↗

    Vals AI · 2026

  49. ProofBench — evaluation methodology and results ↗

    Vals AI · 2026

  50. Public Benefits Bench — evaluation methodology and results ↗

    Vals AI · 2026

  51. Vals RSI Index — evaluation methodology and results ↗

    Vals AI · 2026

  52. SAGE — evaluation methodology and results ↗

    Vals AI · 2026

  53. Vals Science Index — evaluation methodology and results ↗

    Vals AI · 2026

  54. SRE Bench — evaluation methodology and results ↗

    Vals AI · 2026

  55. Tax Agent Bench — evaluation methodology and results ↗

    Vals AI · 2026

  56. Terminal-Bench 4.0 — evaluation methodology and results ↗

    Vals AI · 2026

  57. Terminal-Bench-Science — evaluation methodology and results ↗

    Vals AI · 2026

  58. Vibe Code Bench — evaluation methodology and results ↗

    Vals AI · 2026

Names used in source tables

GPT-6 Sol · GPT-6 Sol (Max) · GPT-6 Sol (Mini-SWE-agent) · GPT-6 Sol (OpenHands) · GPT-6 Sol (high) · GPT-6 Sol (low) · GPT-6 Sol (max) · GPT-6 Sol (medium) · GPT-6 Sol (unknown thinking) · GPT-6 Sol (xhigh) · GPT-6-Sol (max) · gpt-6-sol