Skip to content

chore(release): prepare 0.7.6 #140

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
pierreraby merged 1 commit into from
Oct 6, 2026
Merged

chore(release): prepare 0.7.6 #140

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
6 changes: 5 additions & 1 deletion CHANGELOG.md
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -1,12 +1,16 @@
# Changelog

## Unreleased
## 0.7.6 - 2026-10-06

- Correct the display input price for `stepfun/Step-3.5-Flash` to $0.09 per million tokens and add the `Qwen/Qwen3.6-Plus` long-context tier above 256K tokens, verified against the official pricing page on 2026-10-06. Mark the unchanged, expired `qwen-3.7-max-2x-usage` badge as reviewed in the pricing checker while keeping it visible in reports; changed deal terms and price drift still require review.
- Restore incremental reasoning on the Provider API OpenAI Responses wire by translating Command Code's non-standard `response.reasoning.delta` events into the canonical `response.reasoning_text.delta` before they reach pi-ai, including valid trailing frames without a final blank line. Reasoning now streams while the reasoning item is still open instead of appearing only at `response.output_item.done`. Unsupported reasoning event shapes remain unchanged and produce a content-free warning in `/commandcode-status`. Reproduced with `deepseek/deepseek-v4.1-flash-fast`; standard Responses events, Chat Completions, Anthropic Messages, and the generate fallback are unchanged.
- Add a read-only `npm run check:commandcode-pricing` command that compares the live Provider API catalog and the official pricing page against `MODEL_COSTS`, reporting missing local prices, changed base rates and context tiers, time-policy drift, and expired promotions. It runs in the daily catalog workflow and applicable pull requests, publishes its report to the Actions summary, and never edits prices.
- Add reviewed display pricing for the October catalog additions — `claude-sonnet-5-5`, `gpt-6.1-sol` (with its 272K long-context tier), `deepseek/deepseek-v4.1-flash-fast`, and the free `inclusionai/ling-3.1-flash:free`. Apply the DeepSeek V4 weekday peak-pricing window to `deepseek/deepseek-v4.1-flash-fast`. Models absent from `MODEL_COSTS` silently fall back to a zero display cost, so the snapshot now covers all 85 advertised models.

### Contributors

- @sandexzx — added October catalog display pricing (#135), built the pricing-drift checker (#136), and restored incremental Responses reasoning (#138).

## 0.7.5 - 2026-10-06

- Harden the `test-pi-local.mjs` temp-home cleanup against transient `ENOTEMPTY` when RPC children flush session files while exiting: poll with fresh removal attempts instead of relying on `maxRetries`, whose behavior on `ENOTEMPTY` varies across Node versions. Test-only change; the 0.7.4 release run failed on this cleanup after all tests had passed.
Expand Down
4 changes: 2 additions & 2 deletions package-lock.json
View file
Open in desktop

Some generated files are not rendered by default. Learn more about how customized files appear on GitHub.

2 changes: 1 addition & 1 deletion package.json
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -1,6 +1,6 @@
{
"name": "pi-commandcode-provider",
"version": "0.7.5",
"version": "0.7.6",
"description": "pi custom provider for Command Code API (commandcode.ai)",
"type": "module",
"keywords": [
Expand Down
Loading