No feed. One issue a day. Then you are done.
Weekdays — what shipped, what to skip, what it means. Two minutes on a quiet day, four on a heavy one.
Saturday — one argument, made properly. Sunday — nothing at all.
Prices, limits, context windows, parameter changes and shutdown dates, with the date each one was read. Nobody maintains this systematically, which is exactly why it is here — and it is a by-product of the watchers that feed the daily issue, not extra work.
| Subject | Date | Days left |
|---|---|---|
| claude-sonnet-4-5-20250929 | 2026-09-29 | 54d |
| claude-haiku-4-5-20251001 | 2026-10-15 | 70d |
| claude-opus-4-5-20251101 | 2026-11-24 | 110d |
| Subject | Kind | Was | Now | Effective |
|---|---|---|---|---|
| claude-opus-4-5-20251101 docs.claude.com — model deprecations · read 2026-08-01 | Shutdown | — | Not sooner than 2026-11-24 | 2026-11-24 |
| claude-haiku-4-5-20251001 docs.claude.com — model deprecations · read 2026-08-01 | Shutdown | — | Not sooner than 2026-10-15 | 2026-10-15 |
| claude-sonnet-4-5-20250929 docs.claude.com — model deprecations · read 2026-08-01 | Shutdown | — | Not sooner than 2026-09-29 | 2026-09-29 |
| claude-opus-4-1-20250805 Deprecated 5 June 2026. Confirmed still listed Deprecated with tentative retirement August 5, 2026. Requests after the retirement date fail. docs.claude.com — model deprecations · read 2026-08-05 | Shutdown | — | Retires 2026-08-05 | 2026-08-05 |
| Qwen3.8-Max — context window 2.4T MoE total parameters, 95B active per Alibaba press release. alibabacloud.com — Qwen3.8-Max launch · read 2026-08-05 | Context | — | 1M tokens | 2026-08-03 |
| Qwen3.8-Max — input / output per 1M API generally available 3 Aug 2026 via Alibaba Cloud Model Studio. Open weights promised next week; license not yet disclosed. Alibaba press release does not list token rates; $2/$6 read from qwencloud model page pricing strip and corroborated by multiple reseller pages. qwencloud.com — Qwen3.8-Max · read 2026-08-05 | Price | — | $2.00 / $6.00 | 2026-08-03 |
| temperature, top_p, top_k on Claude Opus 4.7+ Omit the parameter and steer with prompting instead. docs.claude.com — API parameter deprecations · read 2026-08-01 | Parameter | Accepted | Returns 400 on a non-default value | 2026-08-01 |
| deepseek-v4-flash — context window artificialanalysis.ai — DeepSeek V4 Flash 0731 · read 2026-08-01 | Context | — | 1M tokens | 2026-07-31 |
| deepseek-v4-flash — input / output per 1M artificialanalysis.ai — DeepSeek V4 Flash 0731 · read 2026-08-01 | Price | — | $0.14 / $0.28 | 2026-07-31 |
| GPT-5.6 Sol — Fast mode Requests tagged priority continue to work. openai.com — price-performance frontier · read 2026-08-01 | Price | Priority Processing | 2× price, up to 2.5× faster | 2026-07-30 |
| GPT-5.6 Terra — input / output per 1M openai.com — price-performance frontier · read 2026-08-01 | Price | −20% from prior | $2.00 / $12.00 | 2026-07-30 |
| GPT-5.6 Luna — input / output per 1M Announced as an 80% reduction. Prior figure inferred from that percentage, not published directly. openai.com — price-performance frontier · read 2026-08-01 | Price | $1.00 / $6.00 | $0.20 / $1.20 | 2026-07-30 |
| deepseek-chat, deepseek-reasoner Legacy aliases. Verify what your provider resolves them to before assuming they still work. api-docs.deepseek.com — change log · read 2026-08-01 | Shutdown | — | Discontinued 2026-07-24 | 2026-07-24 |
Where a figure is inferred rather than published — a prior price reconstructed from an announced percentage, for instance — the note says so. An inferred number is not a quoted one.
Or Saturdays only, if the week is enough. Full issue in the email — no teaser, no click required.