xAI·released 2026-04-171 source
Grok 4.3 Beta benchmark scores: 10 benchmarks tracked. 40% of its scores are independently reproduced — the rest are self-reported or unverified.
Leads on 0 of 10 benchmarks·best result 93.33 on OTIS Mock AIME 2024-2025 (reproduced)·4 of 10 independently reproduced
| Benchmark | Score | Evidence status | Measured |
|---|---|---|---|
| OTIS Mock AIME 2024-2025 | 93.33accuracy (%) | reproduced· optimizedT1 | 2026-04-17Epoch AI |
| GPQA diamond | 85.1accuracy (%) | reproduced· optimizedT1 | 2026-04-17Epoch AI |
| DTBench | 84.45accuracy (%) | unverifiedT1 | 2026-04-17https://conceptualreasoning.ai/dtbench |
| WeirdML | 49.89accuracy (%) | unverifiedT1 | 2026-04-17https://htihle.github.io/weirdml.html |
| LMCA | 45.09accuracy (%) | unverifiedT1 | 2026-04-17https://conceptualreasoning.ai/lmca |
| FrontierMath-Tiers-1-3-v2-Private | 42.81accuracy (%) | reproducedT1 | 2026-04-17Epoch AI |
| SimpleQA Verified | 33.2accuracy (%) | unverifiedT2 | 2026-04-17 |
| FrontierMath-Tier-4-v2-Private | 14.63accuracy (%) | reproducedT1 | 2026-04-17Epoch AI |
| Chess Puzzles | 13.72accuracy (%) | unverifiedT2 | 2026-04-17 |
| ProofBench | 11accuracy (%) | unverifiedT1 | 2026-04-17https://www.vals.ai/benchmarks/proof_bench |
Claims drawn from cited facts, not live model generation.
This model was developed by xAI and released in 2026.
2 cited facts
This model has scores on 10 tracked benchmarks and currently leads 0 of them. The record’s evidence status is independently reproduced.
3 cited facts
Its mean gap to the best recorded score across its tracked benchmarks is 43.64 points.
1 cited fact
1 facts cross-checked across data sources: 1 single-source.
Source facts, citations, and refresh stamp for this record.
Sources: epoch_benchmarks
Grok 4.3 Beta is an AI model developed by xAI, released 2026-04-17. Its benchmark record below tags every score as independently reproduced, self-reported, or unverified.
Grok 4.3 Beta has recorded scores on 10 benchmarks, each shown with its evidence status.
Grok 4.3 Beta has recorded scores on 10 benchmarks — GPQA diamond, OTIS Mock AIME 2024-2025, DTBench, WeirdML, LMCA, SimpleQA Verified, and 4 more. The full table above shows each score with its evidence status.
4 of 10 recorded scores (40%) are independently reproduced rather than self-reported by the lab.
Grok 4.3 Beta has 10 tracked claims: 4 independently reproduced, 6 unverified.