OpenAI model product
GPT-4.5
5 is OpenAI's most advanced model, offering improved reasoning, coding, and creative capabilities with faster performance and longer context handling than GPT-4.
Updated Aug 17, 2026. Default version: GPT-4.5
LLMBoard Score35.7GPT-4.5
Coverage100%26 benchmark families
Context window128KTokens
Official input priceN/AOfficial price unavailable
GPT-4.5 Capability Profile
This profile uses the model's current scored version. Arena ratings and prices are shown separately.
GPT-4.5 Benchmark Results
Benchmark scores for GPT-4.5.
Sort by
Benchmark | Score | Rank | Participants | Percentile | Evidence | Evaluated |
|---|
| BenchmarkInternal API instruction following (hard) | Score54.0% | Rank02 | Participants7 | Percentile83.3% | EvidenceC | Evaluated |
| BenchmarkSWE-Lancer | Score37.3% | Rank02 | Participants4 | Percentile66.7% | EvidenceC | Evaluated |
| BenchmarkMMLU | Score90.8% | Rank03 | Participants101 | Percentile98.0% | EvidenceC | Evaluated |
| BenchmarkComplexFuncBench | Score63.0% | Rank04 | Participants7 | Percentile50.0% | EvidenceC | Evaluated |
| BenchmarkGraphwalks parents <128k | Score72.6% | Rank04 | Participants11 | Percentile70.0% | EvidenceC | Evaluated |
| BenchmarkGSM8k | Score97.0% | Rank04 | Participants48 | Percentile93.6% | EvidenceC | Evaluated |
| BenchmarkSWE-Lancer (IC-Diamond subset) | Score17.4% | Rank04 | Participants6 | Percentile40.0% | EvidenceC | Evaluated |
| BenchmarkCharXiv-D | Score90.0% | Rank05 | Participants17 | Percentile75.0% | EvidenceC | Evaluated |
| BenchmarkCOLLIE | Score72.3% | Rank05 | Participants10 | Percentile55.6% | EvidenceC | Evaluated |
| BenchmarkGraphwalks BFS <128k | Score72.3% | Rank06 | Participants11 | Percentile50.0% | EvidenceC | Evaluated |
| BenchmarkOpenAI-MRCR: 2 needle 128k | Score38.5% | Rank06 | Participants9 | Percentile37.5% | EvidenceC | Evaluated |
| BenchmarkAider-Polyglot Edit | Score44.9% | Rank07 | Participants10 | Percentile33.3% | EvidenceC | Evaluated |
| BenchmarkSimpleQA | Score62.5% | Rank08 | Participants47 | Percentile84.8% | EvidenceC | Evaluated |
| BenchmarkMathVista | Score72.3% | Rank10 | Participants39 | Percentile76.3% | EvidenceC | Evaluated |
| BenchmarkTAU-bench Airline | Score50.0% | Rank11 | Participants23 | Percentile54.5% | EvidenceC | Evaluated |
| BenchmarkTAU-bench Retail | Score68.4% | Rank13 | Participants25 | Percentile50.0% | EvidenceC | Evaluated |
| BenchmarkMulti-IF | Score70.8% | Rank17 | Participants23 | Percentile27.3% | EvidenceC | Evaluated |
| BenchmarkMMMU | Score75.2% | Rank20 | Participants63 | Percentile69.3% | EvidenceC | Evaluated |
| BenchmarkMulti-Challenge | Score43.8% | Rank20 | Participants29 | Percentile32.1% | EvidenceC | Evaluated |
| BenchmarkIFEval | Score88.2% | Rank26 | Participants67 | Percentile62.1% | EvidenceC | Evaluated |
| BenchmarkHumanEval | Score88.0% | Rank27 | Participants66 | Percentile60.0% | EvidenceC | Evaluated |
| BenchmarkMMMLU | Score85.1% | Rank33 | Participants49 | Percentile33.3% | EvidenceC | Evaluated |
| BenchmarkCharXiv-R | Score55.4% | Rank44 | Participants51 | Percentile14.0% | EvidenceC | Evaluated |
| BenchmarkAIME 2024 | Score36.7% | Rank51 | Participants53 | Percentile3.9% | EvidenceC | Evaluated |
| BenchmarkSWE-Bench Verified | Score38.0% | Rank104 | Participants111 | Percentile6.4% | EvidenceC | Evaluated |
| BenchmarkGPQA | Score69.5% | Rank137 | Participants239 | Percentile42.9% | EvidenceC | Evaluated |
GPT-4.5 Arena Results
Preference and agent-evaluation results for the default version.
Sort by
Arena | Category | Rank | Rating / score | Votes | Observations | Result date |
|---|
| Arenavision style control | Categoryoverall | Rank56 | Rating / score1226.0 | Votes2,925 | ObservationsN/A | Result date |
| Arenatext style control | Categoryoverall | Rank70 | Rating / score1444.7 | Votes14,547 | ObservationsN/A | Result date |
| Arenavision | Categoryoverall | Rank75 | Rating / score1195.1 | Votes2,925 | ObservationsN/A | Result date |
| Arenatext | Categoryoverall | Rank106 | Rating / score1417.4 | Votes14,547 | ObservationsN/A | Result date |
GPT-4.5 Pricing
Official vendor API pricing appears first, followed by individual provider offers.
- Official API
- N/A
- Official provider
- N/A
- Lowest third-party
- N/A
- Tracked offerings
- 0
No provider prices
The default version has no 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.
GPT-4.5 Runtime Performance
Provider-specific output speed and catalog latency for GPT-4.5. Runtime does not affect the capability score.
Sort by
Provider | Output Speed | Catalog Latency | Max Input | Max Output | Updated |
|---|
| ProviderOpenAI | Output Speed50 tok/s | Catalog Latency20 s | Max Input128K | Max Output4.1K | Updated |
Output Speed is generated output tokens received per second. Catalog latency is reported separately from observed provider TTFT.
GPT-4.5 Specifications
Technical details for the model's default version.
- Version
- GPT-4.5
- Released
- Feb 27, 2025
- Knowledge cutoff
- Unknown
- Parameters
- N/A
- Context window
- 128K
- Max output
- 4.1K
- Inputs
- image, text
- Outputs
- text
- Open weights
- No
- License
- Proprietary
GPT-4.5 Versions
Available versions of this model. The score column identifies the version used in the overall ranking.
Sort by
Version | Released | LLMBoard | Parameters | Context | Max output | Open weights | License |
|---|
| VersionGPT-4.5 | Released | LLMBoard35.7 | ParametersN/A | Context128K | Max output4.1K | Open weightsNo | LicenseProprietary |
GPT-4.5 vs nearby models
Open a comparison with the three ranked models immediately above and below this model.
Models similar to GPT-4.5
Recommendations prioritize the same model type and family, then the closest LLMBoard score.
OP
o1
OpenAI
OP
o3 mini
OpenAI
OP
o1 pro
OpenAI
OP
GPT-5.3
OpenAI
OP
GPT-4.1
OpenAI
OP
GPT-5-nano
OpenAI
What is GPT-4.5?
Key information about GPT-4.5 and its available data.
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.
Data as of 2026-08-17.
FAQ
Common questions about GPT-4.5.
When was GPT-4.5 released?
GPT-4.5's default version was released on Feb 27, 2025.
How much does GPT-4.5 cost?
No official standard PAYG price is currently available for GPT-4.5.
Who created GPT-4.5?
GPT-4.5 was created by OpenAI.
What is the context window for GPT-4.5?
The default version has a 128K token context window.
Is GPT-4.5 open weight?
No. The default version is not marked as having publicly available weights.
How many API providers offer GPT-4.5?
No provider offering is currently linked to the default version.
What models should I compare GPT-4.5 with?
Nearby ranked alternatives include LongCat Flash Chat, o1, Qwen3 Next 80B A3B.