OpenAI model product
GPT-4 is a large multimodal model capable of processing both image and text inputs and generating human-like text outputs. It demonstrates human-level performance on various professional and academic benchmarks.
Updated Aug 10, 2026. Default version: GPT-4
Structured fields from the published default version.
This profile uses the latest version under this unique model that has a calculated LLMBoard score. Arena and price are excluded.
Published benchmark records for the scored version GPT-4.
| AI2 Reasoning Challenge (ARC) | 1.0 | 1 | 1 | 100.0% | C | |
| LSAT | 0.9 | 1 | 1 | 100.0% | C | |
| SAT Math | 0.9 | 1 | 1 | 100.0% | C | |
| Uniform Bar Exam | 0.9 | 1 | 1 | 100.0% | C | |
| Winogrande | 0.9 | 1 | 22 | 100.0% | C | |
| HellaSwag | 1.0 | 2 | 27 | 96.2% | C | |
| DROP | 0.8 | 11 | 30 | 65.5% | C | |
| MGSM | 0.7 | 21 | 31 | 33.3% | C | |
| MMLU | 0.9 | 30 | 100 | 70.7% | C | |
| HumanEval | 0.7 | 66 | 76 | 13.3% | C | |
| MATH | 0.4 | 66 | 71 | 7.1% | C | |
| GPQA | 0.4 | 213 | 233 | 8.6% | C |
Preference and agent-evaluation signals from published Arena datasets.
| text style control | polish | 181 | 1315.8 | 179 | N/A | |
| text style control | japanese | 202 | 1201.4 | 1,241 | N/A | |
| text | polish | 205 | 1218.2 | 179 | N/A | |
| text | japanese | 216 | 1114.6 | 1,241 | N/A | |
| text style control | french | 225 | 1269.6 | 1,545 | N/A | |
| text style control | spanish | 228 | 1264.0 | 1,364 | N/A | |
| text style control | german | 231 | 1252.6 | 2,215 | N/A | |
| text style control | korean | 233 | 1177.0 | 1,154 | N/A | |
| text | french | 234 | 1168.7 | 1,545 | N/A | |
| text | korean | 236 | 1060.0 | 1,154 | N/A | |
| text | spanish | 238 | 1167.3 | 1,364 | N/A | |
| text | german | 247 | 1161.4 | 2,215 | N/A | |
| text style control | industry entertainment and sports and media | 248 | 1268.6 | 13,910 | N/A | |
| text style control | math | 248 | 1276.0 | 11,181 | N/A | |
| text style control | industry writing and literature and language | 250 | 1284.0 | 22,099 | N/A | |
| text style control | creative writing | 252 | 1265.7 | 13,932 | N/A | |
| text style control | industry mathematical | 256 | 1271.0 | 9,862 | N/A | |
| text style control | instruction following | 261 | 1275.5 | 29,706 | N/A | |
| text style control | hard prompts | 265 | 1288.3 | 23,040 | N/A | |
| text style control | coding | 266 | 1313.9 | 13,719 | N/A | |
| text style control | hard prompts english | 266 | 1301.0 | 16,244 | N/A | |
| text | math | 270 | 1217.0 | 11,181 | N/A | |
| text style control | expert | 270 | 1250.7 | 3,617 | N/A | |
| text style control | english | 272 | 1292.7 | 57,478 | N/A | |
| text style control | multi turn | 273 | 1261.2 | 13,168 | N/A | |
| text | industry mathematical | 274 | 1201.1 | 9,862 | N/A | |
| text style control | longer query | 274 | 1278.4 | 7,523 | N/A | |
| text style control | russian | 274 | 1256.0 | 5,975 | N/A | |
| text style control | overall | 276 | 1275.3 | 88,723 | N/A | |
| text style control | non english | 276 | 1250.2 | 31,245 | N/A | |
| text style control | exclude ties | 278 | 1185.9 | 60,653 | N/A | |
| text style control | industry medicine and healthcare | 278 | 1263.4 | 4,768 | N/A | |
| text | industry writing and literature and language | 279 | 1214.0 | 22,099 | N/A | |
| text | creative writing | 281 | 1191.8 | 13,932 | N/A | |
| text | industry entertainment and sports and media | 281 | 1180.5 | 13,910 | N/A | |
| text style control | chinese | 281 | 1257.2 | 8,540 | N/A | |
| text style control | industry software and it services | 281 | 1285.8 | 23,796 | N/A | |
| text | instruction following | 284 | 1189.0 | 29,706 | N/A | |
| text style control | industry legal and government | 284 | 1272.4 | 4,695 | N/A | |
| text style control | industry life and physical and social science | 287 | 1275.7 | 15,252 | N/A | |
| text | multi turn | 288 | 1184.7 | 13,168 | N/A | |
| text | russian | 289 | 1172.3 | 5,975 | N/A | |
| text | coding | 290 | 1187.9 | 13,719 | N/A | |
| text | expert | 290 | 1127.2 | 3,617 | N/A | |
| text | hard prompts | 291 | 1174.7 | 23,040 | N/A | |
| text | hard prompts english | 291 | 1190.0 | 16,244 | N/A | |
| text style control | industry business and management and financial operations | 291 | 1231.5 | 8,676 | N/A | |
| text | longer query | 293 | 1188.1 | 7,523 | N/A | |
| text | exclude ties | 294 | 1058.9 | 60,653 | N/A | |
| text | non english | 295 | 1160.3 | 31,245 | N/A | |
| text | chinese | 296 | 1135.2 | 8,540 | N/A | |
| text | overall | 297 | 1186.2 | 88,723 | N/A | |
| text | english | 297 | 1205.5 | 57,478 | N/A | |
| text | industry medicine and healthcare | 298 | 1122.5 | 4,768 | N/A | |
| text | industry legal and government | 299 | 1154.6 | 4,695 | N/A | |
| text | industry software and it services | 301 | 1169.7 | 23,796 | N/A | |
| text | industry business and management and financial operations | 308 | 1121.5 | 8,676 | N/A | |
| text | industry life and physical and social science | 309 | 1153.8 | 15,252 | N/A |
Official vendor API PAYG pricing is summarized first. The table then lists individual provider offerings without treating their minimum as the official price.
| OpenAI | gpt-4 | global | $30 | $60 | 8.2K |
Official prices use only the vendor's configured official Provider and positive standard USD PAYG rates. Third-party offers remain explicitly labeled.
All published versions linked to this unique model. The score columns identify the version used by the current overall ranking.
| GPT-4 | 28.5 | N/A | 8.2K | 8.2K | No | Proprietary |
A concise description based on the published model registry.
GPT-4 is a large multimodal model capable of processing both image and text inputs and generating human-like text outputs. It demonstrates human-level performance on various professional and academic benchmarks.
Use the benchmark, Arena and pricing sections above as separate evidence. A missing field means the current data snapshot does not support that claim.
Data snapshot: 2026-08-07. Editorial model content is not available in the backend.
Open a comparison with the three ranked models immediately above and below this model.
Recommendations prioritize the same model type and family, then the closest published LLMBoard score.
Common questions about GPT-4.
GPT-4's default version was released on Jun 13, 2023.
GPT-4's official API price is $30 per million input tokens and $60 per million output tokens via OpenAI.
GPT-4 is published under OpenAI in the model registry.
The default version has a 8.2K token context window.
No. The default version is not marked as having publicly available weights.
1 published provider offerings are linked to the default version.
Nearby ranked alternatives include Gemini 2.5 Flash Lite, Gemma 2 9B, Llama 3.1 8B.