GPT-4o
Parameters
200B (Estimated)
Context Length
128K
Modality
Multimodal
Architecture
Undisclosed
License
Proprietary
Release Date
13 May 2024
Knowledge Cutoff
Oct 2023
API Pricing (per 1M)
Input: $5.00 · Output: $15.00
Evaluation Benchmarks
Rank
#118
| Benchmark | Score | Rank |
|---|---|---|
Professional Knowledge | 0.747 | 27 |
Software Engineering | 0.39 | 32 |
Graduate-Level QA | 0.701 | 70 |
Coding Index | 0.24 | 113 |
General Text | 1346 | 122 |
Intelligence Index | 0.09 | 178 |
Refactoring Archived | 0.63 | 🥇 1 |
StackEval Archived | 0.961 | 4 |
General Knowledge Reference | 0.887 | 4 |
QA Assistant Archived | 0.956 | 8 |
Summarization Archived | 0.753 | 15 |
Coding Archived | 0.45 | 16 |
Rankings
Overall Rank
#118
Coding Rank
#103
About GPT-4o
GPT-4o is OpenAI's flagship omni-modal model combining text, vision, and audio in a unified architecture. Features real-time responsiveness with superior performance across diverse tasks including reasoning, coding, multilingual understanding, and creative writing. Offers 128K context window with efficient token usage. Represents significant advancement in multimodal AI with seamless integration of different modalities for natural human-computer interaction.
Technical Specifications
Architecture specifications are undisclosed for proprietary models.
Attention
Attention Structure
Multi-Head Attention
Attention Heads
Undisclosed
Key-Value Heads
Undisclosed
Attention Head Dimension
Undisclosed
Position Embedding
Absolute Position Embedding
RoPE Theta
Undisclosed
Sliding Window Attention
Undisclosed
Sliding Window Size
Undisclosed
Sliding Window Ratio
Undisclosed
Linear Attention
Undisclosed
Linear Attention Ratio
Undisclosed
Normalization
Undisclosed
Activation Function
Undisclosed
Dimensions
Hidden Dimension Size
Undisclosed
Number of Layers
Undisclosed
FFN Intermediate Size (Dense)
Undisclosed
Multi-Token Prediction Heads
Undisclosed
Tokenizer
Vocabulary Size
Undisclosed
Resources
About GPT-4o
GPT-4o ("o" for "omni") is OpenAI's flagship multimodal model combining text, vision, and audio understanding in a unified architecture. Features real-time responsiveness, superior multilingual capabilities, and enhanced reasoning. Represents the evolution of the GPT-4 series with improved efficiency and broader modality support.
Other GPT-4o Models
- No related models available