← All lab files

xAI research file

#5 on the current provider board.

Grok 4.20 Beta 1 scores 1474.2 according to LMArena's text style-control snapshot published July 16, 2026.

LMArena source rank#5
LMArena rating1474.2
Reported interval1469.5–1478.8
Battles26,844

Model records from the Epoch AI catalog

xAI 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. Grok 4.20

    LanguageAPI accessClosed weightsRead profile →

  2. Grok 4.1 Fast

    LanguageAPI accessClosed weightsOpen record →

  3. Grok 4.1

    LanguageAPI accessClosed weightsRead profile →

  4. Grok 4 Heavy

    LanguageUnreleasedClosed weightsOpen record →

  5. Grok 4

    Language, Multimodal, VisionAPI accessClosed weightsRead profile →

  6. Grok 3

    Language, Vision, MultimodalAPI accessClosed weightsOpen record →

  7. Grok-2

    Language, Vision, MultimodalAPI accessClosed weightsOpen record →

  8. Grok-1

    LanguageOpen weights (unrestricted)Open weightsOpen record →

Filter every reviewed record in the model explorer →

Historical momentum · legacy metric only

+135.7 from first to last sampled observation.

This change is calculated only inside the selected legacy series, from August 28, 2024 to August 29, 2025. The archive crosses methodology changes and cannot be joined to, trended into, or compared numerically with the current snapshot.

First sample1294.8grok-2-2024-08-13
Legacy peak1430.52025-08-29 · grok-4-0709
Last sample1430.5grok-4-0709
Show all 5 legacy observations
  1. grok-2-2024-08-131294.8
  2. grok-2-2024-08-131289.1
  3. grok-2-2024-08-131287.7
  4. grok-3-preview-02-241399.4
  5. grok-4-07091430.5

Release record

Latest matched notable models.

Epoch AI records provide dated context only. A release record does not imply a score change.

  1. Grok 4.20
  2. Grok 4.1 Fast
  3. Grok 4.1
  4. Grok 4 Heavy
  5. Grok 4

Continue researching

Profiles, guides, and reports.

Evidence boundary

Three sources. Three distinct jobs.

Current frontier: LMArena leaderboard dataset, revision afed939e10281b660a4369206ca505b2bf5e0208, 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