Fable 5 Is Now Available on Filot
Anthropic's newest frontier model, Fable 5, is now available on Filot on an opt-in basis. In our internal evaluations, it delivers a 12% improvement on complex financial modeling, measured on accuracy, comprehensiveness, and instruction adherence.
Today we're making Fable 5, Anthropic's newest frontier model, available on Filot. It's offered on an opt-in basis: nothing changes for your team unless you choose to enable it.
Built for Long-Horizon Work
Complex financial modeling is a long chain of dependent steps, where an error early on compounds through everything downstream. Fable 5 is built for exactly this. In our testing it works 20–30% longer on a task, and that extra time shows up as a markedly more comprehensive result. Filot already ties out every KPI and segment breakout against its validation controls; with Fable 5, that thoroughness comes with even greater consistency — conventions set at the start of a task are still honored hundreds of steps later, and every datapoint ties to its source.
Measured on Our Workflows
We benchmarked Fable 5 on our internal evaluation suite for long-running financial work: tasks like three-statement modeling, DCF and LBO builds, and other complex financial services workflows. Fable 5 and Gemini 3.5 were the strongest frontier models we tested, with Fable 5 delivering a 12% improvement in financial modeling, measured on accuracy, comprehensiveness, and instruction adherence.
Opt-In by Design
Fable 5 is offered as an opt-in option for organizations that want it. It's disabled by default; workspace administrators can enable it from settings, and individual analysts can then select it per workflow. Your existing configuration continues to work exactly as before.
One note on data handling: Anthropic has configured its retention policy slightly differently for this model. Data processed by Mythos-class models is retained for 30 days for trust and safety purposes, with no training on your data — which is why we offer Fable 5 on a customer-elected basis.
"A 12% improvement on financial modeling is the largest single-generation gain we've measured. Where it matters most is Custom Workflows: Fable 5 stays with a task until it's genuinely done, so firms can run their own modeling processes end to end and get results back polished and tied out." — Bjorn Birkelund, Co-Founder of Filot
Want Fable 5 turned on for your firm? Your Filot contact can get you set up.
More from Filot
See all →Fable 5.1 Is Now Available on Filot
Anthropic's newest frontier model is now available on Filot. It leads our internal evaluations on long-running financial modeling tasks, and discounted prompt cache reads make it cheaper to run than Fable 5.
Verifying AI Output Shouldn't Take Longer Than the Work Itself
AI only saves time if you can verify its work efficiently. Filot audits at two levels, deterministic validation checks and granular cell-by-cell traceability, so analysts review AI-generated models at the speed of reading, not the speed of searching.
Claude Sonnet 5 Is Now the Default on Filot
Anthropic's Claude Sonnet 5 is now the default model powering every Filot workflow — stronger reasoning, better tool use, and improved performance on long-running financial modeling tasks.