Use Fable 5 and Fable 5.1 in Harvey

Learn what Fable 5 and Fable 5.1 are, how admins can enable them, and how to use them in the model selector.

Last updated: Sep 18, 2026


Overview

Fable 5 and Fable 5.1 are frontier Mythos-grade models from Anthropic, available in Harvey as premium options. They are built for long-horizon, autonomous work, demonstrating strong performance across legal tasks as measured by Harvey’s benchmarks (LAB and BigLaw Bench).

Because they process your data differently than Harvey's core services, Fable models are governed by Harvey's Covered AI Features framework and are off by default. A workspace admin must opt in before anyone can use them.


Before You Enable Fable 5 or Fable 5.1

There are two key areas to review with your organization before turning Fable 5 on: data retention and regional processing.

Data retention

The use of Fable models, including Fable 5 and Fable 5.1, may involve data processing practices that differ from the data commitments in your existing Harvey agreement.

Category

Details

Retention period

Prompts and outputs are retained for 30 days, then deleted.

Longer retention

Only if content is flagged by a safety classifier, or if Anthropic is legally required to retain it

Human review

No Anthropic employee has default access. A small set of approved reviewers may access content only if it is flagged for potential serious harm, or at your written request. Every access is recorded in a tamper-proof log

Training

Retained data is not used to train Anthropic's models, and is not shared with other customers

Fable model access is isolated to a dedicated Anthropic workspace. Enabling Fable 5 or Fable 5.1 does not affect ZDR for any other model — you can continue using other Anthropic models in Harvey with ZDR enabled.

For the full data-handling detail, see Covered AI Service.

Regional processing

Fable models support US-based processing only. Anthropic does not currently offer regional processing for Fable 5 or Fable 5.1.

This means that if your workspace is hosted outside the US, you can still enable and use Fable models, but every Fable 5 or Fable 5.1 query will be processed in the United States, and stored in the US. Your queries to all other models continue to follow your workspace's normal regional processing configuration.


Enabling and Monitoring Usage of Fable 5 and Fable 5.1 (For Admins)

Enabling Fable 5 and Fable 5.1 is a two-step process.

Step 1: Contact Harvey to Covered AI Services

The Covered AI Services setting is not visible in your workspace until Harvey enables it.

  1. Contact your account team to have Covered AI Services enabled for your workspace.
  2. Your Harvey team will walk you through the data retention and processing terms above, and the commercial terms for Fable model usage.
  3. Once complete, the Covered AI Services toggle appears in your admin settings.

Note: The Covered AI Services setting is subject to Harvey's Service Terms.

Step 2: Turn on Covered AI Services and enable Fable 5

  1. Go to Command Center → Configurations → Models.
  2. Turn on Covered AI Services.
  3. Enable Fable 5 and Fable 5.1 for either your entire workspace or for specific users.

Fable models appear in the model selector only for users you have enabled them for.

Monitoring Usage

You can view your organization's Fable model usage in the Credit Usage page within Settings.

  1. Go to Command Center → Reporting → Credit Usage.
  2. In the Credit usage by model panel, review Fable 5 and Fable 5.1 consumption over time.

Using Fable 5 and Fable 5.1 in Harvey

Fable 5 and Fable 5.1 appear as named models in the model selector, alongside the other models available to you.

In a Harvey Thread:

  1. Select the Settings Slider customize button.
  2. In the models list, select Fable 5 or Fable 5.1.
Harvey's Assistant, with the model selector open. The selected model is Claude Fable 5.

In a Review Table:

  1. Create a new column query and select Continue.
  2. From the model selector, choose Fable 5 or Fable 5.1.

When you select Fable 5 or Fable 5.1 directly, Harvey routes your query to the Fable model, and the model used is shown in your thread. If the Fable model is temporarily unavailable, Harvey displays a notice and automatically routes the query to another model that is well-suited to the task.


FAQs

Was this article helpful?