OpenAI model product
GPT-4
GPT-4 is an OpenAI large multimodal model that processes image and text inputs and generates text outputs.
Updated Sep 8, 2026. Default version: GPT-4
LLMBoard Score4.3GPT-4
Coverage20%8 benchmark families
Context window32.8KTokens
Official input price$30OpenAI API
GPT-4 Capability Profile
This profile uses the model's current scored version. Arena ratings and prices are shown separately.
GPT-4 Benchmark Results
Benchmark scores for GPT-4.
Sort by
Benchmark | Score | Rank | Participants | Percentile | Evidence | Evaluated |
|---|
| BenchmarkAI2 Reasoning Challenge (ARC) | Score96.30% | Rank01 | Participants1 | Percentile100.00% | EvidenceC | Evaluated |
| BenchmarkLSAT | Score88.00% | Rank01 | Participants1 | Percentile100.00% | EvidenceC | Evaluated |
| BenchmarkSAT Math | Score89.00% | Rank01 | Participants1 | Percentile100.00% | EvidenceC | Evaluated |
| BenchmarkUniform Bar Exam | Score90.00% | Rank01 | Participants1 | Percentile100.00% | EvidenceC | Evaluated |
| BenchmarkWinogrande | Score87.50% | Rank01 | Participants22 | Percentile100.00% | EvidenceC | Evaluated |
| BenchmarkHellaSwag | Score95.30% | Rank02 | Participants27 | Percentile96.15% | EvidenceC | Evaluated |
| BenchmarkDROP | Score80.90% | Rank11 | Participants30 | Percentile65.52% | EvidenceC | Evaluated |
| BenchmarkMGSM | Score74.50% | Rank21 | Participants31 | Percentile33.33% | EvidenceC | Evaluated |
| BenchmarkMMLU | Score86.40% | Rank30 | Participants101 | Percentile71.00% | EvidenceC | Evaluated |
| BenchmarkHumanEval | Score67.00% | Rank59 | Participants66 | Percentile10.77% | EvidenceC | Evaluated |
| BenchmarkMATH | Score42.00% | Rank66 | Participants71 | Percentile7.14% | EvidenceC | Evaluated |
| BenchmarkLM Arena Text Style Control | Score1,312.78 rating | Rank176 | Participants210 | Percentile16.27% | EvidenceA | Evaluated |
| BenchmarkLM Arena Text | Score1,263.22 rating | Rank182 | Participants210 | Percentile13.40% | EvidenceA | Evaluated |
| BenchmarkGPQA | Score35.70% | Rank227 | Participants247 | Percentile8.13% | EvidenceC | Evaluated |
GPT-4 Arena Results
Preference and agent-evaluation results for the default version.
Sort by
Arena | Category | Rank | Rating / score | Votes | Observations | Result date |
|---|
| Arenatext style control | Categorypolish | Rank112 | Rating / score1,367.48 | Votes193 | ObservationsN/A | Result date |
| Arenatext | Categorypolish | Rank120 | Rating / score1,326.34 | Votes193 | ObservationsN/A | Result date |
| Arenatext style control | Categoryjapanese | Rank131 | Rating / score1,262.18 | Votes1,422 | ObservationsN/A | Result date |
| Arenatext | Categoryjapanese | Rank140 | Rating / score1,209.31 | Votes1,422 | ObservationsN/A | Result date |
| Arenatext style control | Categoryfrench | Rank146 | Rating / score1,339.57 | Votes1,734 | ObservationsN/A | Result date |
| Arenatext style control | Categorykorean | Rank146 | Rating / score1,252.41 | Votes1,292 | ObservationsN/A | Result date |
| Arenatext | Categoryfrench | Rank149 | Rating / score1,282.37 | Votes1,734 | ObservationsN/A | Result date |
| Arenatext style control | Categorygerman | Rank152 | Rating / score1,297.84 | Votes2,503 | ObservationsN/A | Result date |
| Arenatext style control | Categoryspanish | Rank153 | Rating / score1,311.18 | Votes1,472 | ObservationsN/A | Result date |
| Arenatext | Categorykorean | Rank154 | Rating / score1,183.61 | Votes1,292 | ObservationsN/A | Result date |
| Arenatext | Categorygerman | Rank156 | Rating / score1,250.32 | Votes2,503 | ObservationsN/A | Result date |
| Arenatext | Categoryspanish | Rank157 | Rating / score1,260.16 | Votes1,472 | ObservationsN/A | Result date |
| Arenatext style control | Categoryindustry entertainment and sports and media | Rank160 | Rating / score1,289.81 | Votes15,221 | ObservationsN/A | Result date |
| Arenatext style control | Categorymath | Rank161 | Rating / score1,302.76 | Votes13,306 | ObservationsN/A | Result date |
| Arenatext style control | Categoryindustry mathematical | Rank162 | Rating / score1,308.61 | Votes11,549 | ObservationsN/A | Result date |
| Arenatext style control | Categoryindustry writing and literature and language | Rank165 | Rating / score1,306.70 | Votes24,682 | ObservationsN/A | Result date |
| Arenatext | Categorymath | Rank167 | Rating / score1,269.20 | Votes13,306 | ObservationsN/A | Result date |
| Arenatext | Categoryindustry mathematical | Rank170 | Rating / score1,270.07 | Votes11,549 | ObservationsN/A | Result date |
| Arenatext style control | Categorychinese | Rank170 | Rating / score1,314.08 | Votes9,150 | ObservationsN/A | Result date |
| Arenatext style control | Categorycreative writing | Rank170 | Rating / score1,285.62 | Votes15,576 | ObservationsN/A | Result date |
| Arenatext | Categorychinese | Rank171 | Rating / score1,242.32 | Votes9,150 | ObservationsN/A | Result date |
| Arenatext style control | Categoryinstruction following | Rank172 | Rating / score1,296.01 | Votes34,416 | ObservationsN/A | Result date |
| Arenatext | Categoryindustry entertainment and sports and media | Rank173 | Rating / score1,242.43 | Votes15,221 | ObservationsN/A | Result date |
| Arenatext style control | Categorynon english | Rank173 | Rating / score1,295.26 | Votes35,407 | ObservationsN/A | Result date |
| Arenatext style control | Categorymulti turn | Rank174 | Rating / score1,298.47 | Votes14,867 | ObservationsN/A | Result date |
| Arenatext style control | Categoryrussian | Rank174 | Rating / score1,299.66 | Votes7,053 | ObservationsN/A | Result date |
| Arenatext | Categorycreative writing | Rank175 | Rating / score1,244.10 | Votes15,576 | ObservationsN/A | Result date |
| Arenatext | Categoryindustry writing and literature and language | Rank175 | Rating / score1,264.01 | Votes24,682 | ObservationsN/A | Result date |
| Arenatext style control | Categoryoverall | Rank176 | Rating / score1,312.78 | Votes93,439 | ObservationsN/A | Result date |
| Arenatext style control | Categoryenglish | Rank177 | Rating / score1,325.11 | Votes64,698 | ObservationsN/A | Result date |
GPT-4 Pricing
Official vendor API pricing appears first, followed by individual provider offers.
- Official API
- $30 input, $60 output per 1M
- Official provider
- OpenAI
- Lowest third-party
- N/A
- Tracked offerings
- 1
Sort by
Provider | Provider model ID | Region | Input / 1M | Output / 1M | Context | Updated |
|---|
| ProviderOpenAI | Provider model IDgpt-4 | Regionglobal | Input / 1M$30 | Output / 1M$60 | Context8.2K | Updated |
Official prices use only the vendor's configured official Provider and positive standard USD PAYG rates. Third-party offers remain explicitly labeled.
GPT-4 Runtime Performance
Provider-specific output speed and catalog latency for GPT-4. Runtime does not affect the capability score.
Sort by
Provider | Output Speed | Catalog Latency | Max Input | Max Output | Updated |
|---|
| ProviderAzure | Output Speed104.00 tok/s | Catalog Latency0.30 s | Max Input32.8K | Max Output32.8K | Updated |
| ProviderOpenAI | Output Speed100.00 tok/s | Catalog Latency0.50 s | Max Input32.8K | Max Output32.8K | Updated |
Output Speed is generated output tokens received per second. Catalog latency is reported separately from observed provider TTFT.
GPT-4 Specifications
Technical details for the model's default version.
- Version
- GPT-4
- Released
- Jun 13, 2023
- Knowledge cutoff
- Dec 31, 2022
- Parameters
- N/A
- Context window
- 32.8K
- Max output
- 32.8K
- Inputs
- image, text
- Outputs
- text
- Open weights
- No
- License
- Proprietary
GPT-4 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 | Released | LLMBoard4.25 | ParametersN/A | Context32.8K | Max output32.8K | Open weightsNo | LicenseProprietary |
Models similar to GPT-4
Recommendations prioritize the same model type and family, then the closest LLMBoard score.
OP
GPT-4.1-nano
OpenAI
OP
GPT-4o-mini
OpenAI
OP
GPT-3.5-turbo
OpenAI
OP
GPT-5.1-Codex-Mini
OpenAI
OP
GPT-4-Turbo
OpenAI
OP
o1 mini
OpenAI
What is GPT-4?
Key information about GPT-4 and its available data.
GPT-4 is a large multimodal language model from OpenAI that processes image and text inputs. It generates human-like text outputs and demonstrates human-level performance on various professional and academic benchmarks.
Data as of 2026-09-08.
FAQ
Common questions about GPT-4.
When was GPT-4 released?
GPT-4's default version was released on Jun 13, 2023.
How much does GPT-4 cost?
GPT-4's official API price is $30 per million input tokens and $60 per million output tokens via OpenAI.
Who created GPT-4?
GPT-4 was created by OpenAI.
What is the context window for GPT-4?
The default version has a 32.8K token context window.
Is GPT-4 open weight?
No. The default version is not marked as having publicly available weights.
How many API providers offer GPT-4?
1 provider offerings are linked to the default version.