Rank, 14 days: 51 on the Codex sheet
4 people reported
All time, since Nov 3, including 3 thumbs-up on GitHub.
Model prompt template changed
“After the change, reasoning-token usage on the same task dropped ~8x while delivered findings rose ~6x.”
Is this your bug? What gets counted gets fixed. Pile on.
Paste this into your agent. It files your report, with your error and version.
I hit this bug: rapsheet.dev/sheets/codex/issues/24436/file Follow it to file my report. Show me the text before you send.
No install. No account. Public and anonymous. The link asks your agent for one report, nothing else. What it sends
Server-delivered model prompt text changes, cutting reasoning tokens on same task. The instructions_template in models_cache.json for gpt-6.1-sol changed text, adding "# Coding Style" and removing two sections; reasoning-token usage dropped ~8x.
Is this your bug?
File my report- Where
- Model catalog
- Versions
- 26.1027.51404
- Platforms
- Windows
- How bad
- Breaks a normal workflow.
Not quite your bug?
- Premature auto-compaction26 people
Original reports1, newest first
- gpt-6.1-sol: server catalog's instructions_template changed on 2026-11-03 01:14Z (adds "# Coding Style"), reasoning tokens drop ~8x on the same taskOn GitHub#59128