Grok 4.20 BetavsGrok 4.3 Beta
Grok 4.20 Beta | Grok 4.3 Beta | |
|---|---|---|
| Benchmarks | ||
Nonsense detection BullshitBench v2Given a confidently-worded but nonsensical prompt, does the AI spot that it makes no sense and push back — instead of playing along and inventing an answer? The score is how often it clearly called out the nonsense. Higher is better. | 56%Best | 50% |
Abstract reasoning ARC-AGI-2Puzzle-style tests of abstract reasoning and pattern-finding — the kind of thing people find easy but AIs often struggle with. Higher is better. | 53.3% | — |
Community preference Arena Elo (Text)Real people chat with two anonymous AIs side by side and vote for the answer they prefer. Votes become a chess-style Elo rating on arena.ai — it measures which AI people actually like, not test scores. Higher is better. | 1475 | — |
| Overview | ||
| Company | SpaceXAI | SpaceXAI |
| Release date | Feb 17 2026 | Apr 17 2026 |
| Access | Proprietary | Proprietary |
Which is better: Grok 4.20 Beta or Grok 4.3 Beta?
Grok 4.20 Beta leads Grok 4.3 Beta on 1 of the 1 benchmark they both report (BullshitBench v2). Grok 4.20 Beta shipped 59 days before Grok 4.3 Beta, so benchmark comparisons should account for the intervening progress.
Published specifications for these two models are limited — see each model page for the latest details.
On BullshitBench v2, Grok 4.20 Beta leads at 56% vs Grok 4.3 Beta at 50%.
Frequently asked questions
Grok 4.20 Beta was released by SpaceXAI on Feb 17 2026.
Grok 4.3 Beta was released by SpaceXAI on Apr 17 2026.
Other comparisons
Grok 4.20 BetavsClaude Opus 5Grok 4.3 BetavsClaude Opus 5Grok 4.20 BetavsGPT-5.6 SolGrok 4.3 BetavsGPT-5.6 SolGrok 4.20 BetavsGemini 3.6 FlashGrok 4.3 BetavsGemini 3.6 FlashGrok 4.20 BetavsMuse Spark 1.2Grok 4.3 BetavsMuse Spark 1.2Grok 4.20 BetavsDeepSeek-V4-Flash-0731Grok 4.3 BetavsDeepSeek-V4-Flash-0731Grok 4.20 BetavsMistral Medium 3.5Grok 4.3 BetavsMistral Medium 3.5