ModelsSep 11, 2026 · 21:533 sources

OpenAI Retires 1000 Token Per Second GPT-5.3-Codex-Spark Model Next Week

From huggingnews.com

Image: huggingnews.com
AI;DR

OpenAI will decommission the GPT-5.3-Codex-Spark model next week, which had a peak output of 1,000 tokens per second, as usage has fallen in favor of more advanced iterations. The phase-out marks a shift away from older Codex-Spark variants as the company focuses on a refined ecosystem of assistants and successors.

Read the sourceSign in to suggest

Sources

+ 2 more
  1. SUPPORT@reach_vb“proud of being the first fastest boi out there”x.com
  2. SUPPORT@kimmonismus“5.3-codex-spark was an incredibly fast model, up to 1000 tokens/second”x.com
#openai#gpt#coding#llmMarkdown
Content history (2)
  • 2026-10-16 · Summary · vi · Replaced stored text
    Th. Sottiaux cho biết quyết định ngưng mô hình GPT-5.3-Codex-Spark nhằm tạo chỗ cho các c…
    Th. Sottiaux đã công bố việc ngưng vận hành GPT-5.3-Codex-Spark để nhường chỗ cho các côn…
  • 2026-10-16 · Title · vi · Replaced stored text
    OpenAI sẽ ngừng sử dụng mô hình GPT-5.3-Codex-Spark 1000 token/giây vào tuần tới
    OpenAI ngừng hoạt động mô hình GPT-5.3-Codex-Spark 1000 token/giây tuần tới

One ranked edition, every morning

The stories that matter today. Free, cancel anytime.

Telegram @aidr_today