Skip to content
ClaudeClaude

Claude Fable 5: Working At The Frontier

We asked 5 teams what’s now possible with Claude Fable 5. In this film, teams at Thomson Reuters, Hebbia, Cognition, Cursor, and Base44 describe the before and after in their own words. Working at the Frontier is an ongoing series where the teams building with Claude tell their own story. Learn more about Claude Fable 5: https://www.anthropic.com/claude/fable Office Hours: https://claude.com/office-hours

Jul 6, 20262mWatch on YouTube ↗

At a glance

WHAT IT’S REALLY ABOUT

Customer stories show Fable boosting complex work across industries fast

  1. Hebbia reports about a 20% performance lift on a rigorous finance dataset, suggesting Fable handles complex deal-work tasks that directly impact competitive speed.
  2. Cursor says Fable maintains direction on deep software problems with fewer user nudges, improving reliability for professional coding-agent workflows.
  3. Base44 describes using Fable to regenerate a system prompt in hours instead of days of senior-engineer effort, delivering 90–95% of what was needed.
  4. Thomson Reuters sees growing potential for Claude to help draft high-stakes legal motions that traditionally require extensive context and precision, with quality compounding as models improve.
  5. Cognition (Devin) claims Fable is a step-change for operating inside large companies’ messy internal tool ecosystems, increasing trust for larger projects.

IDEAS WORTH REMEMBERING

5 ideas

Fable expands the set of “approachable” problems for AI users.

Multiple customers describe tasks previously considered too context-heavy or brittle—like legal motions or enterprise toolchains—becoming viable workflows with Fable.

Measured gains in finance translate into competitive advantage.

Hebbia ties model quality directly to compressing the deal life cycle, citing ~20% improvement on a rigorous dataset as evidence of practical impact.

Reliability matters as much as raw coding ability.

Cursor emphasizes reduced need to “repeatedly nudge” the model, framing Fable’s value as staying on track during deep problem-solving rather than requiring constant steering.

Fable can significantly reduce senior-engineering time on prompt/system work.

Base44 contrasts days of top-engineer effort with a four-hour run that produced 90–95% of the needed system prompt, enabling faster iteration and broader product ambition.

High-stakes domains benefit from compounding quality improvements.

Thomson Reuters suggests legal solutions improve as underlying model capability improves, unlocking more precise, context-aware drafting over time.

WORDS WORTH SAVING

5 quotes

There's this whole class of problems that people weren't even necessarily thinking about that are now more approachable with Fable.

Unknown

Previously, I would need to repeatedly nudge the model. With Fable, it really is able to just stay on track.

Cursor

With Fable, it ran for four hours and came back with ninety to ninety-five percent of what we needed.

Base44

This was always a task that required far too much context and precision to be possible. With Claude, we see good potential to do this.

Thomson Reuters

Fable was a step change above any model we'd seen in terms of the ability to survive in these environments.

Cognition

Frontier problem-solving with FableFinance deal-cycle compression and benchmarkingCoding agents staying on trackVibe coding and prompt/system design automationLegal document drafting (motions) and precisionEnterprise internal tools and model robustnessTrust, reliability, and scaling to bigger projects

High quality AI-generated summary created from speaker-labeled transcript.

Get more out of YouTube videos.

High quality summaries for YouTube videos. Accurate transcripts to search & find moments. Powered by ChatGPT & Claude AI.