Steven Gonsalvez

Software Engineer

Codex on AWS bedrock bug causing 10x charges

Why CEREBRO kept it

Codex AWS Bedrock pricing bug alert

The text below is an automated extraction of the article at https://github.com/openai/codex/issues/37674, stored verbatim in the public cerebro-vault repository. Copyright remains with the original publisher (github.com).

Summary Native Codex CLI requests to Amazon Bedrock Mantle cannot opt into GPT-5.6 Sol explicit prompt caching. On an agentic coding workload, this has produced a large volume of cache-write tokens and materially higher cost. This is related to #35300, but adds independent production usage evidence from the native amazon-bedrock provider. Environment - Codex CLI: 0.147.0 - Provider: native amazon-bedrock - Endpoint: Bedrock Mantle Responses API, us-east-1 - Model: openai.gpt-5.6-sol Observed production usage For the completed days 2026-08-05 through 2026-08-08, Cost Explorer usage quantities a

Community take

Bedrock cache writes with <5% hit ratio caused 10x charges; disabling web_search was the fix.

Backlinks

Appeared in 1 briefing

Related

Shares tags: ai/llm-mechanics · cerebro/signal · release-notes

Also from github.com