Update Amazon Bedrock GPT-5.6 context window to 1M tokens

What variant of Codex are you using?

Codex CLI with the built-in Amazon Bedrock provider.

What feature would you like to see?

AWS now advertises a 1 million token context window for OpenAI GPT-5.6 Sol, Terra, and Luna on Amazon Bedrock. Codex currently limits every GPT-5 model in its static Bedrock catalog to 272,000 tokens.

Please update both context_window and max_context_window to 1_000_000 for:

• openai.gpt-5.6-sol
• openai.gpt-5.6-terra
• openai.gpt-5.6-luna

GPT-5.4 and GPT-5.5 should retain their existing 272,000-token limits. context_window controls the active default, while max_context_window caps user configuration overrides, so both fields should reflect the Bedrock limit.

Additional information

AWS announcement: https://aws.amazon.com/about-aws/whats-new/2026/08/gpt-sol-terra-luna-long-context-bedrock/

A tested implementation is available here:

• Commit: https://github.com/zhongkechen/codex/commit/aa0932cdc44cb731966db534f833900144728e7b
• Fork PR: https://github.com/zhongkechen/codex/pull/1

Validation completed with just fmt and just test -p codex-model-provider (59 tests passed).