Skip to main content
Original: Simon Willison · 12/02/2026

​
Summary

an ultra-fast model for real-time coding in Codex OpenAI announced a partnership with Cerebras on January 14th. Four weeks later they’re already launching the first integration, “an ultra-fast model for real-time coding in Codex”.

​
Key Insights

“an ultra-fast model for real-time coding in Codex” — Describing the main feature of GPT-5.3-Codex-Spark.
“at launch, Codex-Spark has a 128k context window and is text-only.” — Clarifying the initial capabilities of Codex-Spark.
“When a model responds this fast you can stay in flow state and iterate with the model much more productively.” — Highlighting the benefits of the model’s speed for coding productivity.

​
Topics

​
Full Article

# Introducing GPT‑5.3‑Codex‑Spark
Author: Simon Willison
Published: 2026-02-12
Source: https://simonwillison.net/2026/Feb/12/codex-spark/#atom-everything
Introducing GPT‑5.3‑Codex‑Spark OpenAI announced a partnership with Cerebras on January 14th. Four weeks later they’re already launching the first integration, “an ultra-fast model for real-time coding in Codex”. Despite being named GPT-5.3-Codex-Spark it’s not purely an accelerated alternative to GPT-5.3-Codex - the blog post calls it “a smaller version of GPT‑5.3-Codex” and clarifies that “at launch, Codex-Spark has a 128k context window and is text-only.” I had some preview access to this model and I can confirm that it’s significantly faster than their other models. Here’s what that speed looks like running in Codex CLI: 
[![
](https://static.simonwillison.net/static/2026/gpt-5.3-codex-spark-medium-last.jpg)](https://static.simonwillison.net/static/2026/gpt-5.3-codex-spark-medium.mp4) That was the “Generate an SVG of a pelican riding a bicycle” prompt - here’s the rendered result: Whimsical flat illustration of an orange duck merged with a bicycle, where the duck's body forms the seat and frame area while its head extends forward over the handlebars, set against a simple light blue sky and green grass background. Compare that to the speed of regular GPT-5.3 Codex medium: 
[![
](https://static.simonwillison.net/static/2026/gpt-5.3-codex-medium-last.jpg)](https://static.simonwillison.net/static/2026/gpt-5.3-codex-medium.mp4) Significantly slower, but the pelican is a lot better: Whimsical flat illustration of a white pelican riding a dark blue bicycle at speed, with motion lines behind it, its long orange beak streaming back in the wind, set against a light blue sky and green grass background. What’s interesting about this model isn’t the quality though, it’s the speed. When a model responds this fast you can stay in flow state and iterate with the model much more productively. I showed a demo of Cerebras running Llama 3.1 70 B at 2,000 tokens/second against Val Town back in October 2024. OpenAI claim 1,000 tokens/second for their new model, and I expect it will prove to be a ferociously useful partner for hands-on iterative coding sessions. It’s not yet clear what the pricing will look like for this new model.

​
Key Takeaways

​
Notable Quotes

an ultra-fast model for real-time coding in Codex
Context: Describing the main feature of GPT-5.3-Codex-Spark.
at launch, Codex-Spark has a 128k context window and is text-only.
Context: Clarifying the initial capabilities of Codex-Spark.
When a model responds this fast you can stay in flow state and iterate with the model much more productively.
Context: Highlighting the benefits of the model’s speed for coding productivity.
  • [[topics/openai-api]]
  • [[topics/prompt-engineering]]
  • [[topics/ai-agents]]

Introducing the Codex app

Simon Willison · explanation · 77% similar

OpenAI Gave Us a Glimpse Into Their AI Coding Playbook

Dan Shipper (Every) · explanation · 71% similar

[AINews] GPT 5.4: SOTA Knowledge Work -and- Coding -and- CUA Model, OpenAI is so very back

Swyx · explanation · 70% similar
Originally published at https://simonwillison.net/2026/Feb/12/codex-spark/#atom-everything.