tegwick
c4f71a1231
Each publication keeps only claims a ledger capture supports. Launch stories, index scores, and Frontier, Efficient, and Budget labels stay excluded. Assistant: grok Assistant-Session: 01a1925d-3496-7011-a33e-6bfb99f4f4d3
1.5 KiB
Price readings — 2026-W41
Read on 31 October 2026 from captures already in the ledger. These lines do not assign Frontier, Efficient, or Budget. They do not replace ledger-2026-09.md or readings-2026-W40.md.
On 6 October, capture 50eb50a66a6c4770b065c379ccd56b96d1f68a7b98e6ed28874780d7cc506421 of https://claude.com/pricing lists Sonnet 5.5 prompt-cache read at $0.20 per million tokens and does not list Haiku 5.5.
On 7 October, capture 16d63ca029d7b392c26fb6f73c5c6dbfe26f65edbceb5c5503caf3f72e0d58d9 lists the following USD per million tokens. The 11 October capture 6f7047207dcfa2c384603d0e71ec6d666789d20ee6d944288e4e74dbb44dbb7e still shows the Haiku 5.5 rows and the Sonnet 5.5 cache-read rate.
| Row | Input | Output | Cache read | Cache write |
|---|---|---|---|---|
| Haiku 5.5, prompts ≤ 100K tokens | $0.10 | $0.50 | $0.01 | $0.125 |
| Haiku 5.5, prompts > 100K tokens | $0.50 | $2.50 | $0.05 | $0.625 |
| Haiku 4.5, legacy table | $1 | $5 | $0.10 | $1.25 |
| Sonnet 5.5 | $2 | $10 | $0.10 | $2.50 |
The page does not state a percentage run-cost reduction. The short-prompt Haiku 5.5 input and output rates are one tenth of the Haiku 4.5 rates on the same page.
OpenRouter lists anthropic/claude-haiku-5.5 from 7 October at prompt 0.0000001 and completion 0.0000005 per token, context 1000000, with a higher override from 100000 prompt tokens. That is a catalog field, not the Anthropic price card. The card is the pricing-page capture above.