v0.0.30-beta
·
13 commits
to main
since this release
GPT-6 Astra support
Codex Pulse now recognizes GPT-6 Astra usage in local rollout logs and includes it in estimated API-equivalent costs across model summaries, timelines, heatmaps, and projections.
- Added standard GPT-6 Astra pricing: $10/M input, $1/M cached input, and $50/M output tokens.
- Supports both the
gpt-6-astraalias and dated Astra model IDs. - Refreshed GPT-5.6 Sol, Terra, and Luna estimates to the latest published API rates.
- Added regression coverage for model matching, cached-input accounting, and fallback pricing.
Cost figures remain estimates based on standard API processing rates. Fast-mode and cache-write charges are not included because those billing dimensions are not present in local rollout usage records.
Full changes: v0.0.29-beta...v0.0.30-beta