GPT-5.4 Model Specs, Costs & Benchmarks (July 2026)
GPT-5.4, developed by OpenAI, features a context window of 1.1M tokens. The model costs $2.50 per million tokens for input and $15.00 per million tokens for output. It was released on March 5, 2026, and has achieved impressive scores in various benchmarks.
Access GPT-5.4 & 210+ other AI models all in one platformTry Magica for free
Overview
 GPT-5.4 | |
|---|---|
Model Provider The organization behind this AI's development | OpenAI |
Input Context Window Maximum input tokens this model can process at once | 1.1M tokens |
Output Token Limit Maximum output tokens this model can generate at once | 128K tokens |
Release Date When this model first became publicly available | March 5, 2026 4 months ago March 5th, 2026 |
Pricing
| GPT-5.4 | |
|---|---|
Short Context Pricing Base rates before a long-context threshold applies | |
Input Token Cost Cost per million input tokens | $2.50 per million tokens |
Output Token Cost Cost per million output tokens | $15.00 per million tokens |
Web Search Cost Additional cost for each web search operation | $0.01 per search |
Cache Read Cost Cost to reuse cached input tokens | $0.25 per million tokens |
Long Context Pricing Rates for prompts with 272,000 or more tokens | |
Input Token Cost Long-context input cost per million tokens | $5.00 per million tokens |
Output Token Cost Long-context output cost per million tokens | $22.50 per million tokens |
Cache Read Cost Long-context cache read cost per million tokens | $0.50 per million tokens |
Capabilities & Features
| GPT-5.4 | |
|---|---|
Input Types Supported input formats | textimagefile |
Output Types Supported output formats | text |
Tokenizer Text encoding system | GPT |
Key Features Advanced capabilities | ✓ Function Calling✓ Structured Output✓ Reasoning Mode✓ Content Moderation |
Reasoning Controls Defaults and configurable reasoning effort | OptionalEffort: Extra High, High, Medium, Low, NoneDefault: Medium |
Benchmarks
| GPT-5.4 | |
|---|---|
Intelligence Index Overall model quality across independent evaluations | 51.4 (Artificial Analysis index; higher is better) |
Coding Index Programming performance across independent evaluations | 71.1 (Artificial Analysis index; higher is better) |
Agentic Index Ability to complete multi-step agentic tasks | 41.1 (Artificial Analysis index; higher is better) |
Best Design Arena Score Highest human-preference Elo score across design arenas | 1,298 Elo (#17 in Game Dev (57.8% win rate)) |
Compare This Model
See how GPT-5.4 compares with other top models
Compare GPT-5.4 with top models in each category:
🚀Programming
Best models for coding and development
🎨Creative & Roleplay
Models optimized for creative writing
📢Marketing
Content creation and marketing tasks
💻Technology
Technical analysis and explanations
🔬Science
Scientific research and analysis
🌐Translation
Multilingual translation tasks