GPT-4o · Nov. 2024
November 2024 checkpoint of gpt-4o with latest improvements.
You always get the exact model you pick — we never silently route you to another.
About GPT-4o · Nov. 2024
The November 2024 checkpoint of GPT-4o is where OpenAI refined its flagship model's most broadly appealing qualities: notably sharper creative writing and meaningfully improved handling of uploaded files and documents. Users consistently praise the combination of speed — 198.3 tokens per second — and cost efficiency that matches GPT-4 Turbo's intelligence at half the price. For interactive applications, chatbots, or creative work that needs to feel natural rather than mechanical, this checkpoint is a well-regarded default choice. That said, it carries a real limitation worth naming: Artificial Analysis scores its reasoning at 17 out of 100 on their Intelligence Index, below the median for comparable models. Complex multi-step reasoning and tasks that demand deep analytical depth are better served elsewhere. The knowledge cutoff of October 2023 is also a practical constraint for anything time-sensitive. Where it earns its place is in production workflows that blend text, image, and document analysis at scale — particularly any creative or customer-facing context where fluency and response speed matter more than cutting-edge reasoning depth.
Best for
- Creative writing and marketing copy that needs to sound natural and human
- Software development assistance, including code generation and debugging (SWE-bench: 33.2%)
- Document and file analysis — spreadsheets, CSVs, PDFs, and business reports
- Customer-facing chatbots and support systems requiring low latency responses
- Multimodal workflows combining text and image inputs for analysis or content generation
Specifications
| Provider | OpenAI |
|---|---|
| Released | 2024-11 |
| Intelligence | Low |
| Speed | Fast |
| Context window | 128,000 tokens |
| Max output | 16,384 tokens |
| Knowledge cutoff | October 2023 |
| Input price | $2.50 / 1M tokens |
| Output price | $10.00 / 1M tokens |
| Request cost | 1 premium request |
| Plan tier | Premium |
| Model ID | gpt-4o-2024-11-20 |
OpenAI retired GPT‑4o inside ChatGPT on February 13, 2026. It remains available through the OpenAI API and here on just4o.chat.
Frequently asked questions
It is a fixed snapshot checkpoint of GPT-4o released on November 20, 2024. The base GPT-4o model launched in May 2024; this checkpoint is an incremental refinement focused on creative writing and file handling, not a major architectural upgrade.
$2.50 per million input tokens and $10.00 per million output tokens at provider list price. That is 50% cheaper than GPT-4 Turbo at comparable quality for most tasks.
128,000 tokens, with a maximum output of 16,384 tokens per response.
Not particularly. Artificial Analysis gives it an Intelligence Index score of 17 out of 100, below the median of 24 for comparable models. For complex multi-step reasoning, a dedicated reasoning model would be a stronger choice.
Text and images. Real-time audio is available via a separate GPT-4o Realtime API offering, not through this model ID. Video input is not currently supported.
The November 2024 checkpoint specifically improved creative writing quality and file analysis depth. Later checkpoints or GPT-4o-mini trade off some capability for lower cost; this version sits at the midpoint between economy and the full frontier.