The Embedding Report.

Anthropic "pauses" token-based billing for its Claude Agent SDK

21d ago · US · primary source: arstechnica.com

Anthropic has paused a billing change for its Claude Agent SDK that would have shifted heavy users to per-token API pricing, the company said Monday, just as the new rates were set to take effect. The change, announced May 13, would have treated SDK usage — including third-party apps and the programmatic “claude -p” command — separately from standard Claude usage via the chat interface or official CLI [1]. Starting June 15, such outside SDK usage would have been billed at Anthropic’s prevailing API rates, with subscribers receiving a monthly usage credit equal to their subscription price [1]. Under the current arrangement, Agent SDK use is limited only by the standard weekly caps applied to a user’s subscription tier [1]. One analysis suggests Claude Opus users begin saving money after just two to three messages per day, and that their subscription could be worth many multiples of its monthly cost in API usage [1]. Anthropic, founded in 2021 by former OpenAI members Daniela and Dario Amodei, has developed the Claude series of large language models with a focus on AI safety [9]. The company was valued at an estimated $965 billion in May 2026, making it the most valuable pure-play AI company in the world [9]. Claude is trained using “constitutional AI,” a technique developed to improve ethical and legal compliance, and since Claude 3 each generation has typically been released in three sizes: Haiku, Sonnet, and Opus [11]. The billing pause arrives during a period of heightened scrutiny for the company. Since January 2026, the U.S. Department of Defense has been in conflict with Anthropic over the use of its products for military purposes and mass domestic surveillance [10]. Federal agencies began phasing out Claude after Anthropic refused to remove contractual prohibitions on those uses, and the DoD designated the company a “supply chain risk,” barring U.S. private military contractors from doing business with the firm [11]. On March 26, 2026, a federal judge issued a temporary injunction against the DoD’s designation [11]. Anthropic did not provide a new timeline for the billing change or indicate whether it would be revised before any future rollout.

product-launch

Background sources we checked (10)
  • arxiv.org ↗ Transformer-based language models have become the default substrate for natural language processing and the pace of new releases has made it hard for practitioners to separate durable ideas from the noise of incremental announcements. This review works at two levels. At the level…
  • arxiv.org ↗ Long-running LLM agents keep valuable state resident on GPUs: KV caches, request schedulers, communication state, and sometimes online adapters. Losing this state after a GPU or communicator failure can discard minutes to hours of work, yet existing recovery mechanisms either res…
  • arxiv.org ↗ Animal collectives navigate cluttered environments through local coordination, yet robot swarms still struggle to reproduce this capability in the physical world. End-to-end learning offers a route to such coordination, but scaling it to embodied swarms remains difficult: standar…
  • arxiv.org ↗ Finance Agent v2 (by Vals AI) has emerged as the reference benchmark for evaluating both Anthropic Claude and OpenAI ChatGPT frontier language models on financial tasks. However, it narrowly deals with periodic reporting from publicly traded companies (SEC 10-K and 10-Q filings),…
  • arxiv.org ↗ Imitation learning for robotic manipulation relies on large sets of human demonstration trajectories, which are often noisy and temporally irregular due to variable operator speed, intermittent pauses, and inconsistent action density. A common preprocessing strategy is time-unifo…
  • arxiv.org ↗ Large language models have become capable reasoners and tool users that write and run code and search the literature, which makes automating the research process itself a realistic goal. We present PAPERCLAW, a harnessed multi-agent system that carries a project autonomously, fro…
  • arxiv.org ↗ Endpoint detection (EPD) is essential for natural turn-taking in streaming speech systems. However, reliably determining the endpoint of an utterance is challenging because speakers often pause mid-utterance due to hesitations and disfluencies. Semantic EPD has emerged as a promi…
  • en.wikipedia.org ↗ Anthropic PBC is an American artificial intelligence (AI) company headquartered in San Francisco, California. It has developed a series of large language models (LLMs) named Claude and has a focus on AI safety. Anthropic was founded in 2021 by former members of OpenAI, including …
  • en.wikipedia.org ↗ Since January 2026, the United States Department of Defense has conflicted with the artificial intelligence company Anthropic over the use of its products for military purposes and mass domestic surveillance.…
  • en.wikipedia.org ↗ Claude is a series of large language models developed by American software company Anthropic. Claude was released as an AI-based chatbot in March 2023. It is also used in AI-assisted software development. Claude is trained using "constitutional AI", a technique developed by Anthr…

Sources

Spot something wrong? Report an issue