Raw SignalOct 6, 20261 min read4 sources

Reported: GPT-6.1 Sol and Astra in Codex may reach about 50 tokens per second

Posts on X and a PC Watch report say GPT-6.1 Sol and GPT-6 Astra are reportedly running about 50% faster, at roughly 50 tokens per second rather than around 30.

In one minute

Posts on X and a PC Watch report say GPT-6.1 Sol and GPT-6 Astra are reportedly running about 50% faster, at roughly 50 tokens per second rather than around 30. According to posts on X, Codex lead Tibo said GPT-6 Astra and GPT-6.1 Sol would get a speed increase of about 50% at default speed. Posts say it applies to

According to posts on X, Codex lead Tibo said GPT-6 Astra and GPT-6.1 Sol would get a speed increase of about 50% at default speed. Posts say it applies to ChatGPT subscribers and supported partner apps, without a settings change.

PC Watch also reported the speed increase. However, posts describe mixed results: one says a later test reached about 35 tokens per second, while others note earlier slowdowns. New Runtime has not found an official statement from OpenAI. An official post or updated documentation describing the speed change would help confirm the report.

Evidence / sources

  1. 1pc.watch.impress.co.jp/docs/news/2145903.html
  2. 2pagewatcher.app
  3. 3x.com/actu_ia/status/2107365241256358299
  4. 4x.com/Angaisb_/status/2107372248365449612
All Raw Signals

Next reads

Read next

Open the graph
  1. 01topicRumorsExplore the rumors topic hub.
  2. 02topicAi ProductsExplore the ai-products topic hub.
  3. 03archiveRaw SignalsOpen every source-linked raw signal.
  4. 04source ledgerSource LedgerInspect the public source evidence graph.
Who read this page?Machine requests, hidden until opened

Loading the privacy-safe route aggregate...

Open the JSON contract