Repository navigation
fix(models): update gpt-5.6 terra/luna pricing - #3326
Conversation
OpenAI cut GPT-5.6 Terra to $2/$12 (cached $0.20) and Luna to $0.20/$1.20 (cached $0.02) per 1M tokens. Long-context tiers and cache-write prices scaled by the unchanged official multipliers (2x input-side, 1.5x output, 1.25x cache write) across the openai, azure, and aws-mantle mappings. https://openai.com/index/advancing-the-price-performance-frontier-with-gpt-5-6/ Claude-Session: https://claude.ai/code/session_014GCx4tQ7GTxcuf92BByQGZ
|
Important Review skippedReview was skipped due to path filters ⛔ Files ignored due to path filters (1)
CodeRabbit blocks several paths by default. You can override this behavior by explicitly including those paths in the path filters. For example, including ⚙️ Run configurationConfiguration used: Repository UI Review profile: CHILL Plan: Pro Plus Run ID: You can disable this status message by setting the Use the checkbox below for a quick retry:
WalkthroughReduced token and cache-write pricing for ChangesGPT-5.6 pricing
Estimated code review effort: 2 (Simple) | ~10 minutes Possibly related PRs
🚥 Pre-merge checks | ✅ 5✅ Passed checks (5 passed)
✨ Finishing Touches🧪 Generate unit tests (beta)
Thanks for using CodeRabbit! It's free for OSS, and your support helps us grow. If you like it, consider giving us a shout-out. Comment |
|
Images automagically compressed by Calibre's image-actions ✨ Compression reduced images by 71%, saving 1.4 MB.
|
|
Images automagically compressed by Calibre's image-actions ✨ Compression reduced images by 7%, saving 42.1 KB.
|
|
Images automagically compressed by Calibre's image-actions ✨ Compression reduced images by 5.6%, saving 31.0 KB.
|
|
Images automagically compressed by Calibre's image-actions ✨ Compression reduced images by 8.6%, saving 45.1 KB.
|
Summary
OpenAI cut GPT-5.6 Terra and Luna pricing (announcement). This updates all six provider mappings (openai, azure, aws-mantle) in
packages/models:$2.50$2.00$0.25$0.20$15.00$12.00$1.00$0.20$0.10$0.02$6.00$1.20Cache-write and over-272K long-context tier prices are rescaled from the new bases using OpenAI's unchanged multipliers (1.25× cache write; 2× input-side / 1.5× output over 272K), as enforced by
openai-gpt56-pricing.spec.ts.Also adds a public changelog entry (
gpt-5-6-terra-luna-price-cut, id 72) announcing the cuts and linking OpenAI's post, with its OG image generated via gpt-image-2.Verification
pnpm build— 17/17 tasks pass (includinguiwith the new changelog entry)openai-gpt56-pricing.spec.ts,validate-model-capabilities.spec.ts, and allpackages/modelsspecs pass (139 tests)pnpm format— clean, no other surfaces hardcode these priceshttps://claude.ai/code/session_014GCx4tQ7GTxcuf92BByQGZ