Gemini 3 Pro Preview (High)
89.8%
niciun interval publicat
aiqbench · Benchmark-uri LLM
Benchmark-uri
- Surse
- 8
- Intrări urmărite
- 22.310
- Ultima sincronizare
- 55 minacum 55 de minute
Toate sursele se sincronizează normal
- Arena · Text· acum 4 ore
- Arena · WebDev· acum 4 ore
- Arena · Vision· acum 4 ore
- Arena · Agent· acum 4 ore
- GPQA Diamond· acum 55 de minute
- SWE-bench· acum 55 de minute
- Terminal-Bench· acum 55 de minute
- ARC-AGI-2· acum 55 de minute
Filtrează după organizație:
Lansat:
Claude Opus 4.5 (Reasoning)
89.5%
niciun interval publicat
Gemini 3 Flash Preview (Reasoning)
89.0%
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 | Gemini 3 Pro Preview (High) | — | — | 89.8% | — | #19–120 | |||
| 2 | Claude Opus 4.5 (Reasoning) | anthropic | nov. 2025 | $5 / $25 | 89.5% | — | #20–62 | ||
| 3 | Gemini 3 Flash Preview (Reasoning) | — | $0.50 / $3 | 89.0% | — | #27–99 | |||
| 4 | Claude 4.1 Opus (Reasoning) | anthropic | aug. 2025 | $15 / $75 | 88.0% | — | — | ||
| 5 | Claude 4.5 Sonnet (Reasoning) | anthropic | — | $3 / $15 | 87.5% | — | — | ||
| 6 | MiniMax-M2.1 | minimax | dec. 2025 | $0.30 / $1.20 | 87.5% | — | #34–201 | ||
| 7 | GPT-5.2 (Xhigh) | openai | dec. 2025 | $1.75 / $14 | 87.4% | — | #34–92 | ||
| 8 | Claude 4 Opus (Reasoning) | anthropic | mai 2025 | $15 / $75 | 87.3% | — | — | ||
| 9 | GPT-5 (High) | openai | aug. 2025 | $1.25 / $10 | 87.1% | — | #64–149 | ||
| 10 | GPT-5.1 (High) | openai | nov. 2025 | $1.25 / $10 | 87.0% | — | #45–242 | ||
| 11 | Grok 4 | xai | — | — | 86.6% | — | — | ||
| 12 | GPT-5 Codex (High) | openai | — | — | 86.5% | — | #17–142 | ||
| 13 | DeepSeek V3.2 Speciale | deepseek | — | — | 86.3% | — | — | ||
| 14 | DeepSeek V3.2 (Reasoning) | deepseek | dec. 2025 | $0.28 / $0.42 | 86.2% | — | #86–247 | ||
| 15 | Gemini 2.5 Pro | iun. 2025 | $1.25 / $10 | 86.2% | — | #60–432 | |||
| 16 | GPT-5.1 Codex (High) | openai | oct. 2026 | $1.25 / $10 | 86.0% | — | #52–375 | ||
| 17 | Gemini 2.5 Pro Preview (Mar' 25) | — | $1.25 / $10 | 85.8% | — | — | |||
| 18 | GLM-4.7 (Reasoning) | zai | dec. 2025 | $0.54 / $1.98 | 85.6% | — | #82–189 | ||
| 19 | Doubao Seed Code | bytedance-seed | — | — | 85.4% | — | — | ||
| 20 | Grok 4.1 Fast (Reasoning) | xai | oct. 2026 | — | 85.4% | — | #138–429 | ||
| 21 | o3 | openai | apr. 2025 | $2 / $8 | 85.3% | — | #79–318 | ||
| 22 | DeepSeek V3.1 (Reasoning) | deepseek | aug. 2025 | — | 85.1% | — | #101–158 | ||
| 23 | DeepSeek V3.1 Terminus (Reasoning) | deepseek | sept. 2025 | $0.27 / $1 | 85.1% | — | #55–168 | ||
| 24 | DeepSeek V3.2 Exp (Reasoning) | deepseek | sept. 2025 | $0.27 / $0.41 | 85.0% | — | #92–334 | ||
| 25 | Grok 4 Fast (Reasoning) | xai | sept. 2025 | — | 85.0% | — | #155–477 | ||
| 26 | Cogito v2.1 (Reasoning) | deepcogito | — | — | 84.9% | — | — | ||
| 27 | DeepSeek R1 0528 (May '25) | deepseek | mai 2025 | $0.50 / $2.15 | 84.9% | — | #88–250 | ||
| 28 | Kimi K2 Thinking | kimi | nov. 2025 | $0.60 / $2.50 | 84.8% | — | #35–209 | ||
| 29 | o1-preview | openai | sept. 2024 | — | 84.8% | — | #147–314 | ||
| 30 | DeepSeek R1 (Jan '25) | deepseek | ian. 2025 | $0.70 / $2.50 | 84.4% | — | #117–389 | ||
| 31 | MiMo-V2-Flash (Reasoning) | xiaomi | dec. 2025 | — | 84.3% | — | #120–321 | ||
| 32 | Qwen3 235B A22B 2507 (Reasoning) | alibaba | iul. 2025 | $0.09 / $0.55 | 84.3% | — | — | ||
| 33 | Claude 4 Sonnet (Reasoning) | anthropic | mai 2025 | $3 / $15 | 84.2% | — | — | ||
| 34 | Gemini 2.5 Flash Preview (Sep '25) (Reasoning) | — | — | 84.2% | — | — | |||
| 35 | o1 | openai | dec. 2024 | $15 / $60 | 84.1% | — | #127–229 |
Se afișează MMLU-Pro (Artificial Analysis) 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.