Claude Fable 5 vs Sonnet 5: Which Should You Use?
.png)
Copy is never one job. There's the headline that carries an entire campaign, and there's the four hundredth product description variant that just needs to be right. Treating them as the same task, with the same model, means either overpaying for routine work, or underpowering the work that matters most.
Anthropic's latest model series is built around that distinction, and it's live in Pencil right now. Fable 5 and Sonnet 5 are available now for copy generation across the platform.
What's New
- Claude Fable 5 is Anthropic's most intelligent generally available model. It's the pick for your highest-stakes copy: the campaign line, the manifesto, the brief where quality is the whole point.
- Claude Sonnet 5 is the new workhorse, with performance close to Anthropic's flagship tier. Anthropic also describes it as its most agentic Sonnet yet, which matters in Pencil: set it as the Conversation Model for your Agents and it plans, reasons, and executes multi-step creative tasks, not just single generations. Sonnet 5 tends to generate text significantly faster than Fable 5, and, as it’s a smaller model, each query uses less electricity. For everyday copywriting at scale, the variants, the adaptations, the always-on production work, it's a strong new default.
The Right Claude for Each Job
The point of a multi-model platform is not having more names in a dropdown. It's matching capability to task. Fable 5 for the copy that carries the campaign. Sonnet 5 for the copy that carries the volume.
That's the Open Garden difference: the best model for the job, the moment it ships. No migration, no rebuild, no waiting for your platform to catch up.
Available Everywhere You Write
Sonnet 5 and Fable 5 are available wherever you generate text in Pencil: in the Editor, in Workflows, and on the Infinite Canvas. Select them as the model directly, or set them as the Conversation Model for your Agents.
Turning them on takes seconds. If your workspace already has an Anthropic model enabled, the new Claude models are switched on automatically. Otherwise, an admin enables them under AI Governance → Settings → Text Generation. Model access stays governed at the workspace level, so teams get the newest capability inside the same controls.
Why This Matters
Model generations are turning over faster than most enterprise procurement cycles. Teams locked to a single provider inherit that provider's release calendar, gaps and all. Teams on Pencil pick from every frontier lab's best work and match it to the task in front of them, with governance, brand controls, and workflows that don't change when the models do.
Frequently asked questions
Claude Fable 5 is Anthropic's most intelligent generally available model. In Pencil, it's available for text and copy generation across the platform.
Fable 5 is Anthropic's top generally available tier, built for the work where quality matters most. Sonnet 5 is the workhorse: performance close to Anthropic's flagship tier, suited to everyday copywriting at scale and to powering Agents. The other Claude models available in Pencil are Haiku - a lite-weight variant of Claude, even faster and more efficient than sonnet, albeit less capable - and Opus, which sits in between Sonnet and Fable on the efficiency-vs- capability continuum.
Use Fable 5 for high-stakes work: top-level strategic brainstorming, creative ideation, campaign lines, manifestos, and briefs where the copy carries the idea. Then, switch to Sonnet 5 when you need speed and scale, and for day-to-day production copy and variants.
No. Both models appear in the same model pickers you already use, and they're enabled automatically in any workspace that already has an Anthropic model switched on.
Articles
.png)
.png)
