<?xml version="1.0" encoding="UTF-8"?><rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom"><channel><title>AI Signal</title><description>Daily dispatch - concise, sourced analysis of the day&apos;s most important AI and tech news.</description><link>https://aisignalhq.com/</link><language>en-us</language><atom:link href="https://aisignalhq.com/rss.xml" rel="self" type="application/rss+xml"/><item><title>Same List Price, Different Bill: Five Pairs of LLM APIs Where the Workload Decides What You Pay</title><link>https://aisignalhq.com/posts/same-list-price-different-bill/</link><guid isPermaLink="true">https://aisignalhq.com/posts/same-list-price-different-bill/</guid><description>Claude Sonnet 5.5 and GPT-6.1 Sol both list at $2 / $10, yet a long-document workload costs $250 on one and $475 on the other. Five pairs, every assumption shown.</description><pubDate>Wed, 30 Sep 2026 00:00:00 GMT</pubDate><category>pricing</category><category>llm-api</category><category>prompt-caching</category><category>comparison</category></item><item><title>Prompt Caching Is Not 10% Everywhere: What a Cache Hit Costs Across Eight Providers</title><link>https://aisignalhq.com/posts/prompt-caching-is-not-10-percent-everywhere/</link><guid isPermaLink="true">https://aisignalhq.com/posts/prompt-caching-is-not-10-percent-everywhere/</guid><description>49 of the 64 current priced models publish a cache-hit price; four providers charge 10% of input. Anthropic, DeepSeek, xAI, Alibaba and Meta differ, with dates.</description><pubDate>Wed, 16 Sep 2026 00:00:00 GMT</pubDate><category>pricing</category><category>prompt-caching</category><category>llm-api</category></item><item><title>The Second Price List: What Batch, Flex, Priority and Data-Sharing Discounts Cost Across Seven Providers</title><link>https://aisignalhq.com/posts/the-second-price-list/</link><guid isPermaLink="true">https://aisignalhq.com/posts/the-second-price-list/</guid><description>Half price to wait, double to skip the queue, 12.5× less input to share data: the second price lists of seven LLM providers, read by hand on September 16, 2026.</description><pubDate>Wed, 16 Sep 2026 00:00:00 GMT</pubDate><category>pricing</category><category>batch-api</category><category>llm-api</category><category>methodology</category></item><item><title>Five Things Decide What You Pay for a Model, and Only One of Them Is the Model</title><link>https://aisignalhq.com/posts/five-things-decide-what-you-pay-for-a-model/</link><guid isPermaLink="true">https://aisignalhq.com/posts/five-things-decide-what-you-pay-for-a-model/</guid><description>Host, time of day, promotions, region and request size move the price of the same model by anything from 1.2× to 10×. Five verified cases, with dates and sources.</description><pubDate>Tue, 15 Sep 2026 00:00:00 GMT</pubDate><category>pricing</category><category>llm-api</category><category>methodology</category></item><item><title>OpenAI&apos;s Pricing Page Has Four Tabs That Look the Same in Plain Text. We Published the Wrong One for Seven Days.</title><link>https://aisignalhq.com/posts/openai-pricing-page-four-tabs-we-published-the-wrong-one-for-seven-days/</link><guid isPermaLink="true">https://aisignalhq.com/posts/openai-pricing-page-four-tabs-we-published-the-wrong-one-for-seven-days/</guid><description>From September 3 to 11, 2026 this site listed GPT-5.6 Sol, Terra and Luna at half their price, read from the Batch tab. What went wrong, what caught it, and what is still not caught.</description><pubDate>Tue, 15 Sep 2026 00:00:00 GMT</pubDate><category>pricing</category><category>openai</category><category>methodology</category><category>corrections</category></item></channel></rss>