Three massive things happened at Anthropic in the span of 72 hours, and if you use Claude for anything — work, coding, research — at least one of them is worth your immediate attention.
**Claude Sonnet 5 dropped, and it's the new default**
Anthropic launched Claude Sonnet 5, which is now the default model in Claude Code, and comes with a native 1-million-token context window.
That context window is enormous — it means you can feed Claude an entire codebase, a years-long email thread, or a book-length research document in a single session without the model losing the thread.
Sonnet 5 is built to be the most agentic Sonnet model yet — it can make plans, use tools like browsers and terminals, and run autonomously at a level that, just a few months ago, required larger and more expensive models.
The pricing is also hard to ignore.
Sonnet 5 is available at an introductory price of $2 per million input tokens and $10 per million output tokens through August 31, 2026, after which it moves to standard pricing of $3 per million input tokens and $15 per million output tokens.
For developers and heavy users, that's a real window to build and experiment cheaply before the price bumps.
Sonnet 5 is available across all plans: it is the default model for Free and Pro plans, and is available to Max, Team, and Enterprise users.
In other words, you already have it.
**Claude Fable 5 is back — with a story behind it**
This one has some genuine drama attached.
Anthropic abruptly shut off Claude Fable 5 and Mythos 5 last month after the Trump administration ordered the company to restrict all foreign nationals, including company employees, from accessing the models.
The reason?
Amazon researchers had found a method of bypassing Fable 5's safeguards — prompting it so that it identified software vulnerabilities, and in one case, the model produced code demonstrating how a vulnerability could be exploited.
After two weeks of government review, that ban is now lifted.
Anthropic said it would begin restoring access to Claude Fable 5 and Mythos 5 after the US Department of Commerce notified the company it had removed its export controls.
Anthropic stated it is redeploying the model with "a new set of classifiers to target and block more cybersecurity tasks."
Whether you find that reassuring or complicated probably depends on how closely you've been watching AI's increasingly tangled relationship with Washington.
**Claude Science: a new product for researchers**
Quietly, in the middle of all this,
Anthropic announced the availability of Claude Science, an application that optimizes the company's large language model for use in scientific laboratories and, especially, within the research operations of pharmaceutical companies.
Claude Science is described as a customizable app that integrates the tools and packages researchers most often use, produces auditable artifacts, and provides flexible access to computing resources.
If you work in life sciences or know someone who does, this one deserves a look.
Three launches. Three days. It's a lot to absorb — but the short version is that Claude just got faster, cheaper, more capable, and available to more people worldwide than it was a week ago. To keep up with how to actually use all of it, https://claudecraft.ca is worth bookmarking.
---
**Sources:**
- https://www.anthropic.com/news/claude-sonnet-5
- https://www.anthropic.com/news
- https://releasebot.io/updates/anthropic
- https://www.aljazeera.com/economy/2026/7/1/us-lifts-restrictions-on-powerful-ai-models-fable-mythos-anthropic-says
- https://9to5google.com/2026/07/01/anthropic-fable-5-returns-to-claude/
- https://www.statnews.com/2026/06/30/anthropic-release-claude-science-ceo-dario-amodei/