Contamination MapText / LLM / Model
GPT-4o
- Organisation
- OpenAI
- Family
- GPT
- Released
- 2024-05-13 (Earliest API snapshot name gpt-4o-2024-05-13 (developers.openai.com/api/docs/models/gpt-4o))
- Training cutoff
- 2023-10-01
Evidence
“GPT-4o's text and voice capabilities were pre-trained using data up to October 2023”
arxiv.org · model card · Section 2 · retrieved 2026-10-15 “Oct 01, 2023 knowledge cutoff”
developers.openai.com · website · Model page header · retrieved 2026-10-15 - Checkpoints
- gpt-4o-2024-05-13, gpt-4o-2024-08-06, gpt-4o-2024-11-20
- Weights
- Closed · proprietary
- Links
- PaperModel card
Training data disclosure: Vague
Level determined for 13% of benchmark cells.
- TestTest data seen0
- SeenSeen, split unclear0
- TrainTraining data seen0
- Related data seen0
- NoNot seen3
- ?Unknown19
“GPT-4o's text and voice capabilities were pre-trained using data up to October 2023, sourced from a wide variety of materials including:”
Other declared training data
-
Select publicly available data (industry-standard machine learning datasets and web crawls) (broad web data: may contain benchmark test sets) · split: unknown
'Industry-standard machine learning datasets' are not named.
Evidence
“Select publicly available data, mostly collected from industry-standard machine learning datasets and web crawls.”
arxiv.org · model card · Section 2 · retrieved 2026-10-15 -
Proprietary data from data partnerships (not yet matched to a dataset) · split: unknown
Evidence
“Proprietary data from data partnerships. We form partnerships to access non-publicly available data, such as pay-walled content, archives, and metadata.”
arxiv.org · model card · Section 2 · retrieved 2026-10-15 -
Web Data (broad web data: may contain benchmark test sets) · split: unknown
Evidence
“Web Data: Data from public web pages provides a rich and diverse range of information”
arxiv.org · model card · Section 2 · retrieved 2026-10-15 -
Code and Math (not yet matched to a dataset) · split: unknown
Evidence
“Including code and math data in training helps the model develop robust reasoning skills”
arxiv.org · model card · Section 2 · retrieved 2026-10-15
Benchmark cells
Classic LLM benchmarks
Unknown MMLU (Measuring Massive Multitask Language Understanding) · training data not disclosed in enough detail
Unknown GSM8K (Grade School Math 8K) · training data not disclosed in enough detail
Unknown MATH (Mathematics Aptitude Test of Heuristics) · training data not disclosed in enough detail
Unknown HumanEval · training data not disclosed in enough detail
Unknown ARC-Challenge (AI2 Reasoning Challenge, Challenge Set) · training data not disclosed in enough detail
Unknown HellaSwag · training data not disclosed in enough detail
Unknown TruthfulQA · training data not disclosed in enough detail
Unknown WinoGrande · training data not disclosed in enough detail
Multimodal LLM benchmarks
Unknown MMMU (Massive Multi-discipline Multimodal Understanding and Reasoning) · training data not disclosed in enough detail
Unknown MMMU-Pro · training data not disclosed in enough detail
Unknown MathVista · training data not disclosed in enough detail
Unknown ChartQA · training data not disclosed in enough detail
Unknown DocVQA · training data not disclosed in enough detail
Unknown AI2D (AI2 Diagrams) · training data not disclosed in enough detail
Not seen CharXiv · its test items were published from 2024-06-26, after the model's stated training-data cutoff (2023-10-01)
Evidence (4)
“GPT-4o's text and voice capabilities were pre-trained using data up to October 2023”
“Oct 01, 2023 knowledge cutoff”
“arXiv:2406.18521v1 [cs.CL] 26 Jun 2024”
“we release all QA pairs for the validation set and keep the answers to the test set private to prevent data leakage.”
Recent LLM benchmarks
Unknown GPQA Diamond · training data not disclosed in enough detail
Unknown MMLU-Pro · training data not disclosed in enough detail
Unknown LiveCodeBench (code generation, release_v6) · training data not disclosed in enough detail
Unknown SWE-bench Verified · training data not disclosed in enough detail
Not seen AIME 2024 · its test items were published from 2024-01-31, after the model's stated training-data cutoff (2023-10-01)
Evidence (5)
“GPT-4o's text and voice capabilities were pre-trained using data up to October 2023”
“Oct 01, 2023 knowledge cutoff”
“2024 AIME I problems and solutions. The test was held on Wednesday, January 31 – Thursday, February 1, 2024.”
“AIME I (main AIME competition date): Thursday, Feb. 1, 2024 between 1:30 pm ET and 5:30 pm ET.”
“AIME II (alternate AIME competition date): Wednesday, Feb. 7, 2024 between 1:30 pm ET and 5:30 pm ET.”
Not seen AIME 2025 · its test items were published from 2025-02-06, after the model's stated training-data cutoff (2023-10-01)
Evidence (4)
“GPT-4o's text and voice capabilities were pre-trained using data up to October 2023”
“Oct 01, 2023 knowledge cutoff”
“2025 AIME I problems and solutions. The test was held on Thursday, February 6, 2025.”
“This year the AIME I will take place on February 6, 2025, while the alternate AIME II will take place on February 12, 2025.”
Unknown MATH-500 · training data not disclosed in enough detail
Notes
No contamination or decontamination statement found in the GPT-4o System Card. Third-party signal on LiveCodeBench (code generation, release_v6) (indirect, not shown as a cell): Performance drops on LeetCode problems released since November 2023 (its stated cutoff). Temporal performance evidence. Accepts image input: “It accepts both text and image inputs, and produces text outputs (including Structured Outputs).” Accepts image input: “GPT-4o[1] is an autoregressive omni model, which accepts as input any combination of text, audio, image, and video and generates any combination of text, audio, and image outputs.”
Sources
- GPT-4o System Card (arXiv 2410.21276v1) · model card · retrieved 2026-10-15
- OpenAI API docs: GPT-4o model page · website · retrieved 2026-10-15