# OpenAI GPT-4

As of October 21, 2026, OpenAI GPT-4 ranks #189 of 213 models overall with 6.7 on the Artificial Analysis Intelligence Index. Its strongest category is coding, where it ranks #122 of 136. OpenAI GPT-4 costs $30.00 per million input tokens and $60.00 per million output tokens. It was released on May 28, 2023.

Page: https://llmbenchmark.ai/models/gpt-4
Data from October 21, 2026. Every figure below names the board it was read from and the day.

- Maker: OpenAI
- Released: May 28, 2023
- Price per million input tokens: $30.00
- Price per million output tokens: $60.00
- Context window: 8K tokens
- Overall rank: 189 of 213, by the Artificial Analysis Intelligence Index
- Artificial Analysis Intelligence Index: 6.7
- Vibe Coding Index: 22.4 (from 1 of 4 coding benchmarks)

Benchmark results (45 of 190):

- SWE-bench Verified: 22.4% (source: SWE-bench (swebench.com), read October 21, 2026)
- SWE-bench Lite: 39.7% (source: SWE-bench (swebench.com), read October 21, 2026)
- SWE-bench (full test set): 12.5% (source: SWE-bench (swebench.com), read October 21, 2026)
- AA Intelligence Index: 6.7 (source: Artificial Analysis (via OpenRouter), read October 21, 2026)
- AA Coding Index: 13.1 (source: Artificial Analysis (via OpenRouter), read October 21, 2026)
- GPQA Diamond (Artificial Analysis run): 34.9% (source: Artificial Analysis, read October 21, 2026)
- MMLU Pro (Artificial Analysis run): 56.2% (source: Artificial Analysis, read October 21, 2026)
- IFBench: 33.2% (source: Artificial Analysis, read October 21, 2026)
- GSM8K: 92.0% (source: Epoch AI benchmarking hub, read October 21, 2026)
- MMLU: 86.4% (source: Epoch AI benchmarking hub, read October 21, 2026)
- DTBench: 62.7% (source: Epoch AI benchmarking hub, read October 21, 2026)
- LMCA: 17.1% (source: Epoch AI benchmarking hub, read October 21, 2026)
- BIG-Bench Hard: 75.1% (source: Epoch AI benchmarking hub, read October 21, 2026)
- AI2 Reasoning Challenge: 96.3% (source: Epoch AI benchmarking hub, read October 21, 2026)
- HellaSwag: 95.3% (source: Epoch AI benchmarking hub, read October 21, 2026)
- WinoGrande: 87.5% (source: Epoch AI benchmarking hub, read October 21, 2026)
- Epoch Capabilities Index: 125.7 (source: Epoch AI benchmarking hub, read October 21, 2026)
- LMArena Text: 1206 Elo (source: LMArena (Hugging Face dataset), read October 21, 2026)
- LMArena Text: coding: 1209 Elo (source: LMArena (Hugging Face dataset), read October 21, 2026)
- LMArena Text: hard prompts: 1200 Elo (source: LMArena (Hugging Face dataset), read October 21, 2026)
- LMArena Text: math: 1230 Elo (source: LMArena (Hugging Face dataset), read October 21, 2026)
- LMArena Text: instruction following: 1201 Elo (source: LMArena (Hugging Face dataset), read October 21, 2026)
- LMArena Text: creative writing: 1191 Elo (source: LMArena (Hugging Face dataset), read October 21, 2026)
- LMArena Text: English: 1220 Elo (source: LMArena (Hugging Face dataset), read October 21, 2026)
- LMArena Text: non-English: 1194 Elo (source: LMArena (Hugging Face dataset), read October 21, 2026)
- LMArena Text: Chinese: 1184 Elo (source: LMArena (Hugging Face dataset), read October 21, 2026)
- LMArena Text: German: 1198 Elo (source: LMArena (Hugging Face dataset), read October 21, 2026)
- LMArena Text: Japanese: 1137 Elo (source: LMArena (Hugging Face dataset), read October 21, 2026)
- LMArena Text: Korean: 1087 Elo (source: LMArena (Hugging Face dataset), read October 21, 2026)
- LMArena Text: French: 1220 Elo (source: LMArena (Hugging Face dataset), read October 21, 2026)
- LMArena Text: Spanish: 1197 Elo (source: LMArena (Hugging Face dataset), read October 21, 2026)
- LMArena Text: Russian: 1195 Elo (source: LMArena (Hugging Face dataset), read October 21, 2026)
- LMArena Text: Polish: 1265 Elo (source: LMArena (Hugging Face dataset), read October 21, 2026)
- LMArena Text: multi-turn: 1206 Elo (source: LMArena (Hugging Face dataset), read October 21, 2026)
- LMArena Text: longer queries: 1190 Elo (source: LMArena (Hugging Face dataset), read October 21, 2026)
- LMArena Text: expert: 1149 Elo (source: LMArena (Hugging Face dataset), read October 21, 2026)
- LMArena Text: hard prompts in English: 1209 Elo (source: LMArena (Hugging Face dataset), read October 21, 2026)
- LMArena Text: software and IT: 1195 Elo (source: LMArena (Hugging Face dataset), read October 21, 2026)
- LMArena Text: mathematical work: 1228 Elo (source: LMArena (Hugging Face dataset), read October 21, 2026)
- LMArena Text: science: 1177 Elo (source: LMArena (Hugging Face dataset), read October 21, 2026)
- LMArena Text: medicine and healthcare: 1145 Elo (source: LMArena (Hugging Face dataset), read October 21, 2026)
- LMArena Text: legal and government: 1191 Elo (source: LMArena (Hugging Face dataset), read October 21, 2026)
- LMArena Text: business and finance: 1145 Elo (source: LMArena (Hugging Face dataset), read October 21, 2026)
- LMArena Text: writing and language: 1219 Elo (source: LMArena (Hugging Face dataset), read October 21, 2026)
- LMArena Text: entertainment and media: 1184 Elo (source: LMArena (Hugging Face dataset), read October 21, 2026)