[OpenAI Releases GPT-5.3-Codex-Spark Research Preview] BlockBeats News, February 13: OpenAI today released the GPT-5.3-Codex-Spark research preview, a streamlined version of GPT-5.3-Codex. It is also OpenAI's first model specifically designed for real-time coding, supporting targeted edits, logic restructuring, or interface optimization, with immediate result visualization. Codex-Spark has been optimized to run on ultra-low latency hardware, enabling near-instantaneous responses—delivering over 1,000 tokens per second while still handling practical coding tasks effectively.