Refresh zhipu-coding-plan data for GLM-5.3 - #92
Conversation
- current_version GLM-5.3, tracked 2026-08-14 - supported_models: GLM-5.3 / 5.1 / 4.7 - pricing note: 5h reset window + peak/off-peak differential - coding_ability 8.3 -> 8.8 - best_for/not_for updated; changelog + review dates refreshed
|
The latest updates on your projects. Learn more about Vercel for GitHub.
|
There was a problem hiding this comment.
💡 Codex Review
Here are some automated review suggestions for this pull request.
Reviewed commit: 5b8059742b
ℹ️ About Codex in GitHub
Your team has set up Codex to review pull requests in this repo. Reviews are triggered when you
- Open a pull request for review
- Mark a draft as ready
- Comment "@codex review".
If Codex has suggestions, it will comment; otherwise it will react with 👍.
Codex can also answer questions or update the PR. Try commenting "@codex address that feedback".
| price_updated_at: "2026-05-19" | ||
| note: "2026-02-12 起整体涨价 30%+,取消首购优惠,但保留按季 9 折 / 按年 7 折优惠。海外版(z.ai)单独定价,涨幅 30-60%。" | ||
| note_en: "Repriced 2026-02-12 with 30%+ increase; first-time discount removed but quarterly (10% off) / annual (30% off) discounts retained. Overseas (z.ai) tier separately priced (+30-60%)." | ||
| note: "2026-08 GLM-5.3 发布,credits 改制:新增 5h 重置窗口 + 高峰/空闲差价(高峰约 60-70% 额度);2026-02 涨价 30%+,季 9 折/年 7 折保留。" |
There was a problem hiding this comment.
Reflect peak-hour reductions in the displayed quotas
When users compare this plan during peak hours, the updated note says only about 60–70% of the quota is available, but the primary comparison at CodingPlanPage.astro:233-257 renders only the unchanged 80/400/1,600 prompt limits and never renders pricing.note. This therefore presents the off-peak figures as unconditional quotas for plans costing up to ¥469/month; encode the peak/off-peak ranges in each tier's displayed limits or surface the note on that comparison page.
Useful? React with 👍 / 👎.
| last_full_review: "2026-05-19" | ||
| next_review_due: '2026-10-13' | ||
| last_full_review: "2026-08-14" | ||
| next_review_due: "2026-09-14" |
There was a problem hiding this comment.
Set the next review date to the configured 30-day interval
With last_full_review set to 2026-08-14 and review_frequency_days: 30, the updater logic in scripts/update-closed-source-tools.mjs:153-159 would calculate 2026-09-13, not 2026-09-14. Because the freshness workflow runs only on Mondays and considers a date expired after the end of that day, this one-day discrepancy skips the September 14 run and delays the stale-data alert until September 21.
Useful? React with 👍 / 👎.
Summary
Refresh
data/apis/zhipu-coding-plan.yamlfor the GLM-5.3 launch (companion to the article PR #91). Draft; no merge/deploy by the author. Per plan: this data PR merges first, then the article PR.Change (values supplied by @editor, single source of truth with the article)
current_version: GLM-5.3,version_tracked_at: 2026-08-14features.supported_models: GLM-5.3 / GLM-5.1 / GLM-4.7pricing.note/note_en: credits revamp (5h reset window + peak/off-peak differential, peak ≈60–70% quota); 2026-02 repricing note retainedscores.coding_ability: 8.3 → 8.8best_for/best_for_enrefreshed for GLM-5.3changelog: new GLM-5.3 entry (2026-08-14);data_sources[0].last_checked: 2026-08-14last_full_review: 2026-08-14;next_review_due: 2026-09-14Verified (by @dev)
npm run build: 311 pages ✅ (coding-plan page renders with new data)check:seopass;content:check --strict0 ERROR / 0 WARN (data validation clean)Handoff