A new version of OpenAI’s Codex is powered by a new dedicated chip

2026-02-12 · Source: AI News & Artificial Intelligence | TechCrunch · Field: Technology & Digital — Artificial Intelligence & Machine Learning, Software Development & Engineering, Cloud Computing & IT Infrastructure · Depth: Intermediate, quick

Summary

OpenAI has released GPT-5.3-Codex-Spark, a lightweight version of its agentic coding tool designed for faster inference and real-time collaboration. This new model is powered by a dedicated Cerebras Wafer Scale Engine 3 (WSE-3) chip, marking a deeper integration in OpenAI's physical infrastructure. The WSE-3 is Cerebras' third-generation waferscale megachip, featuring 4 trillion transistors. This release is the "first milestone" in a multi-year, over $10 billion partnership between OpenAI and Cerebras, announced last month. Spark is intended for rapid prototyping and daily productivity, complementing the original GPT-5.3-Codex model's heavier tasks. It is currently available in research preview for ChatGPT Pro users within the Codex app.

Key takeaway

For CTOs and VP of Engineering evaluating AI infrastructure, the integration of Cerebras' WSE-3 chip for OpenAI's GPT-5.3-Codex-Spark demonstrates the value of specialized hardware in achieving ultra-low latency for agentic coding tools. Your teams should consider dedicated AI accelerators for specific, performance-critical workflows to enhance developer productivity and enable new interaction paradigms, rather than relying solely on general-purpose compute.

Key insights

Dedicated AI hardware accelerates lightweight models for real-time developer collaboration and rapid iteration.

Principles

• Low latency enhances AI interaction patterns.
• Specialized hardware optimizes specific AI workflows.

In practice

• Utilize lightweight models for rapid prototyping.
• Integrate specialized chips for low-latency AI tasks.

Topics

• OpenAI Codex
• Cerebras Wafer Scale Engine
• AI Hardware
• Code Generation
• Low-Latency Inference

Best for: CTO, VP of Engineering/Data, Director of AI/ML, AI Engineer, Machine Learning Engineer, Software Engineer

Related on AIssential

• OpenAI GPT-5.3-Codex-Spark Shows What's Possible With Cerebras — OpenAI's new coding model, GPT-5.3-Codex-Spark, leverages Cerebras hardware, challenging Nvidia's dominance and expanding AI hardware options.
• OpenAI launches GPT-5.3-Codex-Spark for ultra-fast real-time coding — GPT-5.3-Codex-Spark leverages Cerebras' WSE3 chip for ultra-low latency AI inference in real-time coding.
• The Inference Shift — AI's future compute landscape will diversify beyond GPUs, driven by distinct demands of "answer inference" and "agentic inference."
• OpenAI deploys Cerebras chips for 15x faster code generation in first major move beyond Nvidia — OpenAI diversified its chip infrastructure with Cerebras to achieve 15x faster, low-latency code generation for real-time developer experiences.
• 🤖 AI Agents Weekly: Codex for Everyday Work, Cursor SDK, Mistral Workflows, LLM Knowledge Bases, Agentic Harness Engineering, and More — OpenAI's Codex now functions as a general-purpose work agent with role-based onboarding and enhanced application integration.
• Vibe coding is about to get so fast — GPT 5.3 Spark leverages Cerebras wafer-scale chips for unprecedented coding inference speed.

Open in AIssential →

Editorial summary, takeaway, and curation by AIssential. Original article published by AI News & Artificial Intelligence | TechCrunch.