GPT-5.4
openai.com
March 5, 2026
17 min read
🔥🔥🔥🔥🔥
75/100
Summary
GPT-5.4, now available in ChatGPT and the API, is the most capable and efficient frontier model designed for professional work. It combines advanced reasoning, coding capabilities from GPT-5.3-Codex, and enhanced agentic workflows for improved performance on complex tasks.
Key Takeaways
- OpenAI released GPT-5.4, the most capable and efficient model for professional work, enhancing reasoning, coding, and agentic workflows.
- GPT-5.4 achieves a new state of the art on the GDPval benchmark, matching or exceeding industry professionals in 83.0% of comparisons.
- The model supports up to 1 million tokens of context, allowing for complex workflows and improved task execution across applications.
- GPT-5.4 shows significant improvements in creating and editing spreadsheets and presentations, with mean scores of 87.3% and 68.0% preference from human raters, respectively.
Coverage & discussions
Hacker News · openai.com
Community discussions
Hacker News discussion ↗
Historical count: 5 · 5 saved in sample · Historical sample
Analysis ready
Community Sentiment
Positives
- GPT-5.4's 1M context window significantly exceeds the ~200k tokens supported by other models, potentially enabling more complex interactions and improved performance in extensive tasks.
- The pricing for GPT-5.4 is competitive, being much cheaper than Opus 4.6, which could make advanced AI more accessible for developers.
Concerns
- The reliance on coordinate-based clicking and screenshot interpretation in GPT-5.4 raises concerns about efficiency and practicality compared to using APIs directly.
- The confusion around versioning and model availability suggests a lack of clarity and stability in OpenAI's offerings, which could frustrate developers.
- Skepticism remains regarding the actual benefits of the 1M context window, as previous models like Codex and Opus have shown weaknesses despite their capabilities.
