Anthropic × New Model Releases

Anthropic launches Fable, Mythos 5.1

Anthropic launches Fable, Mythos 5.1

✎ Story body

A model launch, a usage-limit reset, a Salesforce tie-up and a municipal deployment all landed in the same week, shifting attention from the model itself to how it gets delivered.

What happened

Anthropic announced Claude Fable 5.1 and Claude Mythos 5.1, disclosing concerns about the difficulty of oversight alongside the release. Around it came a blanket reset of usage limits, "Claudeforce" with Salesforce, and a Japanese city running citizen services on Claude — five items from trade press and community sources in one week.

Why it matters

Read together, the emphasis sits less on model capability than on distribution and oversight: risk disclosed at launch, quotas rebuilt, and entry into business systems and public-sector front desks. The delivery layer moved at the same pace as the model. Evidence of how far adoption actually reaches remains thin for now.

What to watch

Whether 5.1 usage settles after the quota reset, and whether Claudeforce moves from announcement to documented deployments. Also worth tracking is how the oversight caveat is made concrete in later operating guidance.

▲ Official & Press
Press

Anthropic、「Claude Fable 5.1」と「Claude Mythos 5.1」発表 「監視の難しさ」への懸念も開示

ITmedia AI+ ・ 2026-09-01 ・ 📌

Anthropic unveils Claude Fable 5.1 and Mythos 5.1, notes new risks

Press

茨城県鹿嶋市は「公式Web×Claude」で市民対応をどう変えた? Claude Codeで内製化も

ITmedia AI+ ・ 2026-09-02

Kashima City runs a Claude chatbot built on its official website data

Community

Claude's new system prompt really doesn't want to reproduce song lyrics

Simon Willison's Weblog ・ 2026-09-02

Claude's Fable 5.1 prompt adds bans on lyrics and characters

Community

Quoting Rick Brewster

Simon Willison's Weblog ・ 2026-09-02

Paint.NET gets a 180K-line, Claude-written Direct2D rewrite for WINE

Press

ClaudeにSalesforceを統合した「Claudeforce」、AnthropicとSalesforceが発表。Claudeから営業データ分析や顧客対応を実現

Publickey ・ 2026-09-02

Anthropic and Salesforce unveil Claudeforce integration

Press

「Claude」利用制限を“一括リセット” 「Fable 5.1」のリリースに伴い

ITmedia AI+ ・ 2026-09-02

Anthropic resets Claude usage limits alongside Fable 5.1 launch

Community

Claude Fable 5.1 made me a really nice animated pelican

Simon Willison's Weblog ・ 2026-09-01

Simon Willison on Fable 5.1: 52.6% science bench, five reasoning levels

Community

Claude Fable 5.1 and Claude Mythos 5.1

Hacker News (Front Page) ・ 2026-09-01

Anthropic releases Claude Fable 5.1 and Mythos 5.1, ~25% cheaper

Academic (arxiv etc.) 2 ▾
Academic

Defense-as-Skill: Evolving Runtime Guard Skill for Skill-Augmented Agents

arXiv cs.AI (Artificial Intelligence) ・ 2026-09-01

Academic

What Does an Agentic Software Engineering Benchmark Measure? Profiling Task Demands and Agent Behaviour Beyond What Category Labels Reveal

arXiv cs.CL (Computation and Language) ・ 2026-09-01

← Story Archive