OpenAI model product
GPT-4.5 is OpenAI's most advanced model, offering improved reasoning, coding, and creative capabilities with faster performance and longer context handling than GPT-4. It features enhanced instruction following, reduced hallucinations, and better factual accuracy.
Updated Aug 10, 2026. Default version: GPT-4.5
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-4.5.
| Internal API instruction following (hard) | 0.5 | 2 | 7 | 83.3% | C | |
| SWE-Lancer | 0.4 | 2 | 4 | 66.7% | C | |
| MMLU | 0.9 | 3 | 100 | 98.0% | C | |
| ComplexFuncBench | 0.6 | 4 | 7 | 50.0% | C | |
| GSM8k | 1.0 | 4 | 48 | 93.6% | C | |
| SWE-Lancer (IC-Diamond subset) | 0.2 | 4 | 6 | 40.0% | C | |
| CharXiv-D | 0.9 | 5 | 16 | 73.3% | C | |
| COLLIE | 0.7 | 5 | 10 | 55.6% | C | |
| Graphwalks parents <128k | 0.7 | 6 | 18 | 70.6% | C | |
| OpenAI-MRCR: 2 needle 128k | 0.4 | 6 | 9 | 37.5% | C | |
| Aider-Polyglot Edit | 0.4 | 7 | 10 | 33.3% | C | |
| SimpleQA | 0.6 | 8 | 46 | 84.4% | C | |
| Graphwalks BFS <128k | 0.7 | 10 | 22 | 57.1% | C | |
| MathVista | 0.7 | 10 | 39 | 76.3% | C | |
| TAU-bench Airline | 0.5 | 11 | 23 | 54.5% | C | |
| TAU-bench Retail | 0.7 | 13 | 25 | 50.0% | C | |
| Multi-IF | 0.7 | 16 | 20 | 21.1% | C | |
| MMMU | 0.8 | 20 | 63 | 69.3% | C | |
| Multi-Challenge | 0.4 | 20 | 29 | 32.1% | C | |
| IFEval | 0.9 | 26 | 65 | 60.9% | C | |
| HumanEval | 0.9 | 29 | 76 | 62.7% | C | |
| MMMLU | 0.9 | 33 | 49 | 33.3% | C | |
| CharXiv-R | 0.6 | 40 | 47 | 15.2% | C | |
| AIME 2024 | 0.4 | 51 | 53 | 3.9% | C | |
| SWE-Bench Verified | 0.4 | 97 | 104 | 6.8% | C | |
| GPQA | 0.7 | 131 | 233 | 44.0% | C |
Preference and agent-evaluation signals from published Arena datasets.
The default version has no published Arena rows, or its source alias has not been resolved.
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-4.5 | 58.8 | N/A | 128K | 4.1K | No | Proprietary |
A concise description based on the published model registry.
GPT-4.5 is OpenAI's most advanced model, offering improved reasoning, coding, and creative capabilities with faster performance and longer context handling than GPT-4. It features enhanced instruction following, reduced hallucinations, and better factual accuracy.
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-4.5.
GPT-4.5's default version was released on Feb 27, 2025.
No official standard PAYG price is currently available for GPT-4.5.
GPT-4.5 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 Kimi K2 Thinking, Nemotron 3 Super, Claude Opus 3.