Knowledge Evaluation on IKP
71.9Accuracy (IKP)GPT-5.5
Evaluation Results
| Method | Links | |
|---|---|---|
| GPT-5.5Vendor=OpenAI, Est. Size=~9.7T, 90% PI=[3.2–28.7T]2026.04 | 71.9 | |
| Claude Opus 4.6Vendor=Anthropic, Est. Size=~5.3T, 90% PI=[1.8–15.6T]2026.04 | 68 | |
| GPT-5 ProVendor=OpenAI, Est. Size=~4.1T, 90% PI=[1.4–12.2T]2026.04 | 66.5 | |
| GPT-5Vendor=OpenAI, Est. Size=~4.1T, 90% PI=[1.4–12.1T]2026.04 | 66.4 | |
| Claude Opus 4.7Vendor=Anthropic, Est. Size=~4.0T, 90% PI=[1.4–12.0T]2026.04 | 66.4 | |
| o1Vendor=OpenAI, Est. Size=~3.5T, 90% PI=[1.2–10.3T]2026.04 | 65.4 | |
| Claude Opus 4.5Vendor=Anthropic, Est. Size=~3.4T, 90% PI=[1.1–10.0T]2026.04 | 65.2 | |
| Claude Opus 4.1Vendor=Anthropic, Est. Size=~3.2T, 90% PI=[1.1–9.5T]2026.04 | 64.9 | |
| Grok-4Vendor=xAI, Est. Size=~3.2T, 90% PI=[1.1–9.4T]2026.04 | 64.8 | |
| o3Vendor=OpenAI, Est. Size=~3.0T, 90% PI=[1.0–8.9T]2026.04 | 64.4 | |
| GPT-5.4 ProVendor=OpenAI, Est. Size=~2.2T, 90% PI=[736B–6.5T]2026.04 | 62.5 | |
| GPT-4.1Vendor=OpenAI, Est. Size=~2.2T, 90% PI=[719B–6.4T]2026.04 | 62.3 | |
| Grok-3Vendor=xAI, Est. Size=~2.1T, 90% PI=[715B–6.3T]2026.04 | 62.3 | |
| Claude Sonnet 4.6Vendor=Anthropic, Est. Size=~1.7T, 90% PI=[579B–5.1T]2026.04 | 60.9 | |
| GPT-5.3Vendor=OpenAI, Est. Size=~1.5T, 90% PI=[503B–4.5T]2026.04 | 60 | |
| GPT-5.2 ProVendor=OpenAI, Est. Size=~1.4T, 90% PI=[478B–4.2T]2026.04 | 59.7 | |
| Claude Opus 4Vendor=Anthropic, Est. Size=~1.4T, 90% PI=[478B–4.2T]2026.04 | 59.7 | |
| GPT-5.1Vendor=OpenAI, Est. Size=~1.3T, 90% PI=[450B–4.0T]2026.04 | 59.3 | |
| GPT-5.2Vendor=OpenAI, Est. Size=~1.3T, 90% PI=[417B–3.8T]2026.04 | 58.9 | |
| Gemini 2.5 ProVendor=Google, Est. Size=~1.2T, 90% PI=[387B–3.4T]2026.04 | 58.4 | |
| GPT-5.4Vendor=OpenAI, Est. Size=~1.0T, 90% PI=[348B–3.1T]2026.04 | 57.7 | |
| GPT-4oVendor=OpenAI, Est. Size=~720B, 90% PI=[241B–2.1T]2026.04 | 55.3 | |
| Qwen3-MaxVendor=Alibaba, Est. Size=~685B, 90% PI=[229B–2.0T]2026.04 | 55 | |
| GPT-4Vendor=OpenAI, Est. Size=~666B, 90% PI=[223B–2.0T]2026.04 | 54.8 | |
| GPT-4-TurboVendor=OpenAI, Est. Size=~630B, 90% PI=[211B–1.9T]2026.04 | 54.5 | |
| GPT-5 MiniVendor=OpenAI, Est. Size=~410B, 90% PI=[137B–1.2T]2026.04 | 51.7 | |
| Gemini 2.5 FlashVendor=Google, Est. Size=~207B, 90% PI=[69B–617B]2026.04 | 47.4 | |
| Claude 3.5 HaikuVendor=Anthropic, Est. Size=~158B, 90% PI=[53B–470B]2026.04 | 45.6 | |
| GPT-5 NanoVendor=OpenAI, Est. Size=~71B, 90% PI=[24B–212B]2026.04 | 40.5 | |
| Claude Haiku 4.5Vendor=Anthropic, Est. Size=~65B, 90% PI=[22B–194B]2026.04 | 39.9 |