gpt-6-astra
95.0%
niciun interval publicat
Română · beta. Traducerea a fost făcută automat și este încă în curs de revizuire. Citește pagina în engleză
aiqbench · Benchmark-uri LLM
95.0%
niciun interval publicat
94.2%
niciun interval publicat
93.3%
niciun interval publicat
O singură grilă, toate coloanele. Alege axe din panoul Axe — o axă clasifică precum un singur clasament, mai multe se combină într-un scor compus ponderat, toate la un loc se citesc ca indexul. Locul este atribuit înaintea oricărui filtru și nu se renumerotează niciodată.
| Model | Organizație | ± CI | Oscilație | Adaugă la comparație | |||||
|---|---|---|---|---|---|---|---|---|---|
| 1 | gpt-6-astra | OpenAI | sept. 2026 | $10 / $50 | 95.0% | — | #1–18 | ||
| 2 | gpt-6.1-sol | OpenAI | oct. 2026 | $2 / $10 | 94.2% | — | #1–15 | ||
| 3 | claude-opus-5-5 | Anthropic | oct. 2026 | $4 / $20 | 93.3% | — | #2–11 | ||
| 4 | gpt-5.6-sol | OpenAI | sept. 2026 | $4 / $20 | 92.5% | — | #6–24 | ||
| 5 | claude-opus-5 | Anthropic | sept. 2026 | $5 / $25 | 90.4% | — | #5–11 | ||
| 6 | claude-fable-5-1 | Anthropic | sept. 2026 | $10 / $50 | 90.0% | — | #2–22 | ||
| 7 | gpt-6-sol | OpenAI | oct. 2026 | $2 / $10 | 89.6% | — | #7–53 | ||
| 8 | claude-fable-5 | Anthropic | sept. 2026 | $10 / $50 | 89.2% | — | #4–21 | ||
| 9 | gemini-3.8-flash | Google DeepMind | oct. 2026 | $0.83 / $4.13 | 89.2% | — | #2–30 | ||
| 10 | gpt-5.5 | OpenAI | sept. 2026 | $5 / $30 | 85.0% | — | #9–34 | ||
| 11 | gemini-3.7-flash | Google DeepMind | oct. 2026 | $0.75 / $3.75 | 84.6% | — | #9–57 | ||
| 12 | gemini-3-deep-think-preview | Google DeepMind | — | — | 84.6% | — | #84–163 | ||
| 13 | gpt-5.5-pro | OpenAI | — | $30 / $180 | 84.6% | — | #59–85 | ||
| 14 | gpt-5.6-terra | OpenAI | sept. 2026 | $2 / $12 | 83.9% | — | #13–60 | ||
| 15 | gpt-5.4-pro-2026-03-05 | OpenAI | mar. 2026 | $30 / $180 | 83.3% | — | #20–171 | ||
| 16 | gemini-3.1-pro-preview | Google DeepMind | sept. 2026 | $2 / $12 | 77.1% | — | #9–36 | ||
| 17 | claude-opus-4-7 | Anthropic | sept. 2026 | $5 / $25 | 75.8% | — | #1–18 | ||
| 18 | gpt-5.4-2026-03-05 | OpenAI | sept. 2026 | $2.50 / $15 | 74.0% | — | #12–27 | ||
| 19 | deepseek-v4.1-flash | DeepSeek | oct. 2026 | $0.15 / $0.60 | 72.9% | — | #14–30 | ||
| 20 | claude-opus-4-8 | Anthropic | sept. 2026 | $5 / $25 | 72.1% | — | #15–33 | ||
| 21 | gemini-3.5-flash | Google DeepMind | sept. 2026 | $1.50 / $9 | 72.1% | — | #10–26 | ||
| 22 | claude-opus-4-6 | Anthropic | sept. 2026 | $5 / $25 | 69.2% | — | #2–25 | ||
| 23 | grok-4.6 | xAI | sept. 2026 | $2 / $6 | 67.1% | — | #15–52 | ||
| 24 | glm-5.3-flash | Z.ai (Zhipu AI) | aug. 2026 | $0.05 / $0.20 | 65.8% | — | #26–59 | ||
| 25 | grok-4-20 | xAI | — | $1.25 / $2.50 | 65.1% | — | #44–259 | ||
| 26 | deepseek-v4-flash-0731 | DeepSeek | iul. 2026 | $0.15 / $0.15 | 61.4% | — | #45–244 | ||
| 27 | grok-4.7 | xAI | oct. 2026 | $2 / $6 | 61.4% | — | #23–72 | ||
| 28 | deepseek-v4-pro-0813 | DeepSeek | aug. 2026 | $0.66 / $1.98 | 61.3% | — | #40–130 | ||
| 29 | claude-sonnet-4-6 | Anthropic | sept. 2026 | $3 / $15 | 60.4% | — | #39–78 | ||
| 30 | gemini-3.6-flash | Google DeepMind | sept. 2026 | $0.75 / $3.75 | 60.4% | — | #16–95 | ||
| 31 | kimi-k3 | Moonshot | iun. 2026 | $0.24 / $11.94 | 60.4% | — | #10–40 | ||
| 32 | gpt-5.6-luna | OpenAI | sept. 2026 | $0.20 / $1.20 | 59.5% | — | #30–147 | ||
| 33 | gpt-6-luna | OpenAI | oct. 2026 | $0.10 / $0.50 | 59.3% | — | #31–85 | ||
| 34 | gpt-5.2-pro-2025-12-11 | OpenAI | dec. 2025 | $21 / $168 | 54.2% | — | #172–262 | ||
| 35 | gpt-5.2-2025-12-11 | OpenAI | dec. 2025 | $1.75 / $14 | 52.9% | — | #34–93 |
Se afișează ARC-AGI-2 exact așa cum a publicat-o sursa — propriul scor și interval de încredere, fără mix. Selectează o a doua axă ca să construiești un scor compus ponderat.