OpenAI model product
GPT-4o ('o' for 'omni') is a multimodal AI model that accepts text, audio, image, and video inputs, and generates text, audio, and image outputs. It matches GPT-4 Turbo performance on text and code, with improvements in non-English languages, vision, and audio understanding.
Updated Aug 10, 2026. Default version: GPT-4o
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-4o.
| ActivityNet | 0.6 | 1 | 1 | 100.0% | C | |
| ComplexFuncBench | 0.7 | 1 | 7 | 100.0% | C | |
| AI2D | 0.9 | 3 | 32 | 93.5% | C | |
| EgoSchema | 0.7 | 3 | 9 | 75.0% | C | |
| SWE-Lancer | 0.3 | 3 | 4 | 33.3% | C | |
| SWE-Lancer (IC-Diamond subset) | 0.1 | 5 | 6 | 20.0% | C | |
| Internal API instruction following (hard) | 0.3 | 7 | 7 | 0.0% | C | |
| COLLIE | 0.6 | 8 | 10 | 22.2% | C | |
| OpenAI-MRCR: 2 needle 128k | 0.3 | 8 | 9 | 12.5% | C | |
| Aider-Polyglot Edit | 0.2 | 9 | 10 | 11.1% | C | |
| ChartQA | 0.9 | 12 | 24 | 52.2% | C | |
| CharXiv-D | 0.9 | 12 | 16 | 26.7% | C | |
| Graphwalks parents <128k | 0.4 | 13 | 18 | 29.4% | C | |
| DocVQA | 0.9 | 15 | 26 | 44.0% | C | |
| Graphwalks BFS <128k | 0.4 | 17 | 22 | 23.8% | C | |
| Multi-IF | 0.6 | 19 | 20 | 5.3% | C | |
| TAU-bench Airline | 0.4 | 19 | 23 | 18.2% | C | |
| Aider-Polyglot | 0.3 | 20 | 22 | 9.5% | C | |
| TAU-bench Retail | 0.6 | 20 | 25 | 20.8% | C | |
| SimpleQA | 0.4 | 21 | 46 | 55.6% | C | |
| Tau2 Airline | 0.5 | 21 | 23 | 9.1% | C | |
| Multi-Challenge | 0.4 | 22 | 29 | 25.0% | C | |
| ERQA | 0.4 | 23 | 23 | 0.0% | C | |
| Tau2 Retail | 0.6 | 23 | 26 | 12.0% | C | |
| VideoMMMU | 0.6 | 25 | 26 | 4.0% | C | |
| MathVista | 0.6 | 26 | 39 | 34.2% | C | |
| MMMU | 0.7 | 28 | 63 | 56.5% | C | |
| Tau2 Telecom | 0.2 | 34 | 35 | 2.9% | C | |
| CharXiv-R | 0.6 | 36 | 47 | 23.9% | C | |
| MMLU | 0.9 | 37 | 100 | 63.6% | C | |
| MMMLU | 0.8 | 38 | 49 | 22.9% | C | |
| MMMU-Pro | 0.6 | 50 | 65 | 23.4% | C | |
| IFEval | 0.8 | 51 | 65 | 21.9% | C | |
| AIME 2024 | 0.1 | 53 | 53 | 0.0% | C | |
| MMLU-Pro | 0.7 | 72 | 129 | 44.5% | C | |
| Humanity's Last Exam | 0.1 | 87 | 92 | 5.5% | C | |
| SWE-Bench Verified | 0.3 | 99 | 104 | 4.8% | C | |
| GPQA | 0.7 | 127 | 233 | 45.7% | C |
Preference and agent-evaluation signals from published Arena datasets.
| vision style control | chinese | 91 | 1072.5 | 186 | N/A | |
| vision | chinese | 93 | 1004.4 | 186 | N/A | |
| vision style control | overall | 112 | 1119.4 | 3,376 | N/A | |
| vision style control | english | 112 | 1122.5 | 1,744 | N/A | |
| vision | overall | 116 | 1064.7 | 3,376 | N/A | |
| vision | english | 118 | 1073.1 | 1,744 | N/A | |
| text style control | japanese | 150 | 1291.9 | 1,104 | N/A | |
| text | japanese | 164 | 1234.6 | 1,104 | N/A | |
| text style control | korean | 167 | 1288.9 | 712 | N/A | |
| text style control | creative writing | 181 | 1326.7 | 6,756 | N/A | |
| text style control | industry entertainment and sports and media | 181 | 1324.9 | 7,043 | N/A | |
| text style control | industry writing and literature and language | 182 | 1340.4 | 12,147 | N/A | |
| text style control | german | 183 | 1313.1 | 1,202 | N/A | |
| text style control | spanish | 186 | 1325.9 | 546 | N/A | |
| text | korean | 188 | 1218.4 | 712 | N/A | |
| text | spanish | 199 | 1268.2 | 546 | N/A | |
| text style control | french | 199 | 1326.9 | 457 | N/A | |
| text | german | 201 | 1257.2 | 1,202 | N/A | |
| text style control | industry legal and government | 204 | 1347.5 | 2,674 | N/A | |
| text style control | russian | 206 | 1323.8 | 5,517 | N/A | |
| text | french | 208 | 1261.4 | 457 | N/A | |
| text style control | non english | 208 | 1315.9 | 20,213 | N/A | |
| text style control | instruction following | 209 | 1318.3 | 18,305 | N/A | |
| text style control | longer query | 209 | 1338.3 | 6,160 | N/A | |
| text style control | multi turn | 212 | 1326.6 | 8,201 | N/A | |
| text style control | overall | 213 | 1335.3 | 45,499 | N/A | |
| text style control | exclude ties | 213 | 1286.0 | 28,538 | N/A | |
| text | creative writing | 214 | 1274.9 | 6,756 | N/A | |
| text style control | industry life and physical and social science | 214 | 1348.1 | 7,560 | N/A | |
| text style control | math | 214 | 1308.6 | 6,826 | N/A | |
| text | industry writing and literature and language | 215 | 1289.6 | 12,147 | N/A | |
| text style control | industry mathematical | 215 | 1312.8 | 5,805 | N/A | |
| text style control | industry medicine and healthcare | 215 | 1342.0 | 2,174 | N/A | |
| text | industry entertainment and sports and media | 217 | 1268.0 | 7,043 | N/A | |
| text style control | english | 217 | 1348.2 | 25,286 | N/A | |
| text | math | 219 | 1285.4 | 6,826 | N/A | |
| text style control | chinese | 220 | 1329.2 | 3,778 | N/A | |
| text style control | expert | 221 | 1309.9 | 2,349 | N/A | |
| text | industry mathematical | 222 | 1283.2 | 5,805 | N/A | |
| text | russian | 225 | 1271.0 | 5,517 | N/A | |
| text | instruction following | 228 | 1267.0 | 18,305 | N/A | |
| text style control | coding | 228 | 1360.4 | 7,318 | N/A | |
| text style control | hard prompts english | 230 | 1339.1 | 8,010 | N/A | |
| text | longer query | 231 | 1283.8 | 6,160 | N/A | |
| text | expert | 232 | 1239.9 | 2,349 | N/A | |
| text style control | hard prompts | 232 | 1326.8 | 12,655 | N/A | |
| text | chinese | 233 | 1253.2 | 3,778 | N/A | |
| text | multi turn | 233 | 1278.8 | 8,201 | N/A | |
| text | non english | 235 | 1261.4 | 20,213 | N/A | |
| text style control | industry software and it services | 235 | 1347.2 | 11,864 | N/A | |
| text | overall | 236 | 1282.7 | 45,499 | N/A | |
| text | exclude ties | 238 | 1209.5 | 28,538 | N/A | |
| text | english | 241 | 1297.6 | 25,286 | N/A | |
| text style control | industry business and management and financial operations | 241 | 1302.2 | 5,095 | N/A | |
| text | coding | 242 | 1284.0 | 7,318 | N/A | |
| text | industry legal and government | 242 | 1275.3 | 2,674 | N/A | |
| text | industry medicine and healthcare | 242 | 1251.1 | 2,174 | N/A | |
| text | hard prompts | 244 | 1264.3 | 12,655 | N/A | |
| text | hard prompts english | 245 | 1278.7 | 8,010 | N/A | |
| text | industry life and physical and social science | 245 | 1274.4 | 7,560 | N/A | |
| text | industry software and it services | 247 | 1278.6 | 11,864 | N/A | |
| text | industry business and management and financial operations | 252 | 1231.5 | 5,095 | 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.
The default version has no provider offering with current input or output token prices.
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-4o | 27.9 | N/A | 128K | 16.4K | No | Proprietary | |
| ChatGPT-4o Latest | N/A | N/A | 128K | 16.4K | No | Proprietary | |
| GPT-4o | N/A | N/A | 128K | 4.1K | No | Proprietary |
A concise description based on the published model registry.
GPT-4o ('o' for 'omni') is a multimodal AI model that accepts text, audio, image, and video inputs, and generates text, audio, and image outputs. It matches GPT-4 Turbo performance on text and code, with improvements in non-English languages, vision, and audio understanding.
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-4o.
GPT-4o's default version was released on Aug 6, 2024.
No official standard PAYG price is currently available for GPT-4o.
GPT-4o is published under OpenAI in the model registry.
The default version has a 128K token context window.
No. The default version is not marked as having publicly available weights.
No published provider offering is currently linked to the default version.
Nearby ranked alternatives include GPT-4, GPT-4.1, Qwen3.5 4B.