Source URL: https://simonwillison.net/2025/Oct/6/gpt-5-pro/
Source: Simon Willison’s Weblog
Title: GPT-5 pro
Feedly Summary: GPT-5 pro
Here’s OpenAI’s model documentation for their GPT-5 pro model, released to their API today at their DevDay event.
It has similar base characteristics to GPT-5: both share a September 30, 2024 knowledge cutoff and 400,000 context limit.
GPT-5 pro has maximum output tokens 272,000 max, an increase from 128,000 for GPT-5.
As our most advanced reasoning model, GPT-5 pro defaults to (and only supports) reasoning.effort: high
It’s only available via OpenAI’s Responses API. My LLM tool doesn’t support that in core yet, but the llm-openai-plugin plugin does. I released llm-openai-plugin 0.7 adding support for the new model, then ran this:
llm install -U llm-openai-plugin
llm -m openai/gpt-5-pro “Generate an SVG of a pelican riding a bicycle"
It’s very, very slow. The model took 6 minutes 8 seconds to respond and charged me for 16 input and 9,205 output tokens. At $15/million input and $120/million output this pelican cost me $1.10!
Here’s the full transcript. It looks visually pretty simpler to the much, much cheaper result I got from GPT-5.
Tags: ai, openai, generative-ai, llms, llm-pricing, pelican-riding-a-bicycle, llm-reasoning, llm-release, gpt-5
AI Summary and Description: Yes
Summary: The text discusses the release of OpenAI’s GPT-5 pro model, noting its enhanced capabilities compared to its predecessor, GPT-5. This information is particularly pertinent to professionals involved in AI development and security, as it highlights advancements in AI reasoning capabilities, token limits, and pricing structure, which could have implications for cost management and usage policies.
Detailed Description: The announcement regarding OpenAI’s GPT-5 pro model provides several key insights into its specifications and operational details. The content is extremely relevant for professionals in AI, generative AI security, and those involved in cloud computing due to its potential impact on model deployment and security practices. Here are the major points:
– **Model Characteristics**:
– GPT-5 pro shares a knowledge cutoff date of September 30, 2024, and maintains a context limit of 400,000 tokens, similar to GPT-5.
– Maximum output tokens for GPT-5 pro is significantly increased to 272,000 from 128,000 in GPT-5, enhancing its capability for generating extensive and complex outputs.
– **Reasoning and Performance**:
– GPT-5 pro is designated as OpenAI’s most advanced reasoning model, which emphasizes a high-effort default mode for reasoning. This shift indicates potential improvements in decision-making tasks and problem-solving, making it a significant upgrade for AI applications that demand sophistication.
– **Availability and Deployment**:
– The model is only accessible through OpenAI’s Responses API, which necessitates developers to adapt their tools to engage with the new model effectively.
– The mention of the llm-openai-plugin enhancing support for the model indicates a need for professionals to stay updated with available tools that can interface with the latest models.
– **Performance and Cost**:
– The author documents a practical experiment with the GPT-5 pro, noting that the model response times were lengthy, taking over six minutes for a simple request. This is essential for businesses to consider when planning for real-time processing requirements.
– The cost incurred for generating a simple output was $1.10, raising points about budget management in AI operations, particularly with varying pricing strategies for input and output tokens.
– **Practical Implications**:
– The advancements in token limits and reasoning capabilities present both opportunities and challenges for organizations utilizing AI models, especially in security contexts.
– Understanding token pricing is critical for firms trying to manage AI costs effectively, as misuse or overuse can lead to significant financial implications.
Overall, the release of GPT-5 pro signifies important advancements for practitioners in AI and infrastructure security, particularly regarding performance expectations, financial management, and operational integration challenges associated with adopting new technologies in existing infrastructures.