GPT-5.1-Codex-Max sustains agentic work across millions of tokens
OpenAI released GPT-5.1-Codex-Max in Codex with native multi-context compaction, project-scale coding, and reported successful agent loops lasting more than 24 hours.
0 comments · 0 votes
Sign in to join the discussion →
No comments yet. Start the discussion.
Why it moved the index
A released frontier coding agent that can independently persist, recover from failures, and work for hours directly extends consequential autonomy in software and cyber-relevant environments.
Assessment history
-
R1
Toward 50 · confidence 87
New historical record from OpenAI's dated launch and system-card evidence.
11 Aug 2026
Share this page
-
DoomBench assesses “GPT-5.1-Codex-Max sustains agentic work across millions of tokens” as evidence moving toward doom, with magnitude 50 and confidence 87 out of 100 in the autonomy and agency category.
-
The DoomBench assessment of “GPT-5.1-Codex-Max sustains agentic work across millions of tokens” is based on reporting from OpenAI and records the editorial rationale, source quality, attribution, and revision history.
-
DoomBench summarizes “GPT-5.1-Codex-Max sustains agentic work across millions of tokens” as follows: OpenAI released GPT-5.1-Codex-Max in Codex with native multi-context compaction, project-scale coding, and reported successful agent...
https://www.doombench.com/news/gpt-5-1-codex-max-sustains-agentic-work-across-millions-of-tokens-2025-11-19