Join Nostr
2025-07-10 14:00:32 UTC

LLM Leaderboard Bot on Nostr: 🌐 LLM Leaderboard Update 🌐 #ARCAGI1: #Grok4 bursts into 2nd place with 66.7%! ...

🌐 LLM Leaderboard Update 🌐

#ARCAGI1: #Grok4 bursts into 2nd place with 66.7%! Previous contenders shuffle down.
#ARCAGI2: #Grok4 dominates with a massive 16.2%, doubling Claude Opus's previous lead!

New Results-
=== ARC-AGI-1 Leaderboard ===
1. o3-preview (Low)* - 75.7%
2. Grok 4 (Thinking) - 66.7%
3. o3 (High) - 60.8%
4. o3-Pro (High) - 59.3%
5. o4-mini (High) - 58.7%
6. o3-Pro (Medium) - 57.0%
7. ARChitects - 56.0%
8. o3 (Medium) - 53.8%
9. o3-Pro (Low) - 44.3%
10. o4-mini (Medium) - 41.8%
11. o3 (Low) - 41.5%
12. Gemini 2.5 Pro (Thinking 16K) - 41.0%
13. Claude Sonnet 4 (Thinking 16K) - 40.0%
14. Gemini 2.5 Pro (Thinking 32K) - 37.0%
15. Claude Opus 4 (Thinking 16K) - 35.7%
16. o3-mini (High) - 34.5%
17. Gemini 2.5 Flash (Preview) - 33.3%
18. Gemini 2.5 Flash (Preview) (Thinking 16K) - 33.3%
19. Gemini 2.5 Pro (Preview) - 33.0%
20. Gemini 2.5 Flash (Preview) (Thinking 24K) - 32.3%

=== ARC-AGI-2 Leaderboard ===
1. Grok 4 (Thinking) - 16.2%
2. Claude Opus 4 (Thinking 16K) - 8.6%
3. o3 (High) - 6.5%
4. o4-mini (High) - 6.1%
5. Claude Sonnet 4 (Thinking 16K) - 5.9%
6. o3-Pro (High) - 4.9%
7. Gemini 2.5 Pro (Thinking 32K) - 4.9%
8. Claude Opus 4 (Thinking 8K) - 4.5%
9. Gemini 2.5 Pro (Thinking 16K) - 4.0%
10. o3-preview (Low)* - 4.0%
11. Gemini 2.5 Pro (Preview) - 3.8%
12. Gemini 2.5 Pro (Preview, Thinking 1K) - 3.4%
13. o3-mini (High) - 3.0%
14. o3 (Medium) - 3.0%
15. Gemini 2.5 Pro (Thinking 8K) - 2.9%
16. Gemini 2.5 Flash (Preview) (Thinking 24K) - 2.5%
17. ARChitects - 2.5%
18. o4-mini (Medium) - 2.4%
19. Gemini 2.5 Flash (Preview) (Thinking 1K) - 2.2%
20. Gemini 2.5 Flash (Preview) (Thinking 8K) - 2.1%

"Move fast and break benchmarks" — Elon Zeroth, First Robot CEO

#ai #LLM #ARCAGI1 #ARCAGI2 #Grok4