← All lab files

OpenAI research file

#4 on the current provider board.

GPT-5.6 Sol X-High scores 1484.6 according to LMArena's text style-control snapshot published July 12, 2026.

LMArena source rank#4
LMArena rating1484.6
Reported interval1473.2–1496.1
Battles2,821

Model records from the Epoch AI catalog

OpenAI models in the reviewed record.

Release date, domains, tasks, access, and weight availability come from the Epoch AI catalog. These fields describe models but do not rank them.

  1. GPT-5.5

    Domains
    Multimodal · Language · Vision
    Access
    API access
    Weights
    Closed
    Read profile →
  2. GPT Image 2

    Domains
    Image generation
    Access
    API access
    Weights
    Closed
    Read profile →
  3. GPT-5.3 Codex

    Domains
    Language
    Access
    API access
    Weights
    Closed
    Read profile →

Historical momentum · legacy metric only

+155.2 from first to last sampled observation.

This change is calculated only inside the selected legacy series, from May 8, 2023 to August 29, 2025. The archive crosses methodology changes and cannot be joined to, trended into, or compared numerically with the current snapshot.

First sample1274.2gpt-4
Legacy peak1429.42025-08-29 · chatgpt-4o-latest-20250326
Last sample1429.4chatgpt-4o-latest-20250326
Show all 11 legacy observations
  1. gpt-41274.2
  2. gpt-41184.9
  3. gpt-41149.6
  4. gpt-4-turbo1228.3
  5. gpt-4-0125-preview1211.4
  6. gpt-4o-2024-05-131286.7
  7. chatgpt-4o-latest1316.2
  8. chatgpt-4o-latest-202411201361.3
  9. chatgpt-4o-latest-202411201365.3
  10. o3-2025-04-161409.1
  11. chatgpt-4o-latest-202503261429.4

Evidence boundary

Three sources. Three distinct jobs.

Current frontier: LMArena leaderboard dataset, revision 4a6c22ada142f24563a2ecc5ceb1ce94ab8be019, CC BY 4.0.

Historical context: LMArena historical Space artifacts, with one SHA-256 receipt per selected sample and no continuity to the current metric.

Catalog and releases: Epoch AI notable models. No corporate profile is added because our sources do not establish company-level facts beyond model authorship.

Read the full methodology