Artificial intelligence news.
Straight from the original announcements.

Claude Fable 5.2 arrives on Vercel’s AI Gateway

November 9, 2026Source: Vercel
Claude Fable 5.2 arrives on Vercel’s AI Gateway
Image: Vercel

Anthropic’s Claude Fable 5.2 is now available through Vercel’s AI Gateway, giving developers access to a model designed for open-ended technical work. The model supports coding, reasoning and vision tasks, alongside orchestration and advisory roles for other models.

Key points
  • Claude Fable 5.2 is now available on Vercel AI Gateway.
  • The model targets open-ended technical work across coding, reasoning and vision.
  • Prompts and completions are retained for 30 days and are not used to train Claude.
  • Zero Data Retention is not supported for Claude Fable 5.2.
  • AI Gateway supports fallbacks to other models when a classifier declines a request.

Built for extended technical work

According to Vercel, Fable 5.2 improves on earlier Fable models in coding, reasoning and vision. It can run long, ambiguous tasks unattended, which positions it for technical workflows that may require several steps or sustained problem solving.

The model can also work as an orchestrator, handing simpler tasks to smaller models. Alternatively, it can act as an advisor that other Claude models consult when they encounter harder problems. Vercel says Fable 5.2 communicates more clearly and generates output faster than earlier Fable models.

Developers can set the model’s effort from low to max, with high set as the API default.

Safety controls and retention requirements

Claude Fable 5.2 has cybersecurity and biology safety classifiers enabled. Access requires data retention, and Anthropic does not support Zero Data Retention for the model.

Prompts and completions are retained for 30 days and are not used to train Claude, according to the announcement. If a team has Zero Data Retention enabled, AI Gateway returns an error instead of sending the request to Fable 5.2.

Vercel says teams can use model fallbacks to keep requests served when a classifier is triggered. By adding a models array to providerOptions.gateway, AI Gateway tries the listed models in order when the cybersecurity classifier declines a request.

Access through multiple APIs and tools

Developers can call the model through the AI SDK, Chat Completions, Responses and Anthropic Messages APIs, as well as a coding agent connected to AI Gateway. The model name is anthropic/claude-fable-5.2.

Vercel’s example uses Opus 5.5 followed by Sonnet 5.5 as fallbacks. In that configuration, a request falls back to Opus 5.5 and then Sonnet 5.5 if the cybersecurity classifier declines it.

Claude Fable 5.2 can also be used in Claude Code, Codex, Cursor and more after installing the latest Vercel CLI and running npx vercel ai-gateway setup. Vercel also provides a model playground for testing it.

What this means for AI Gateway users

The addition gives teams a single route to use Fable 5.2 while configuring retries, failover and performance optimizations through AI Gateway. The service also tracks usage and cost and includes custom reporting, budgets for API keys and routing rules.

Vercel says AI Gateway reflects provider pricing with no markup and does not charge a platform fee on inference, including Bring Your Own Key requests. The main limitation for teams using Fable 5.2 is the retention requirement, since Zero Data Retention requests are rejected rather than routed to the model.

Related news

Gemini 4 Carbon arrives on Vercel’s AI Gateway at 50% off
ModelsNovember 10, 2026

Gemini 4 Carbon arrives on Vercel’s AI Gateway at 50% off

Gemini 4 Carbon from Google is now available on Vercel’s AI Gateway, with access offered at 50% off through November 30, 2026. The model is designed for complex software…

Vercel Adds DeepSeek V4.1 Flash 1017 Ultrafast to AI Gateway
Products and toolsNovember 7, 2026

Vercel Adds DeepSeek V4.1 Flash 1017 Ultrafast to AI Gateway

Vercel has added DeepSeek V4.1 Flash 1017 Ultrafast to its AI Gateway through Fireworks, giving developers access to a model that can generate upwards of 1,000 tokens…

Mistral Large 4 gets updated weights and lower prices on Vercel AI Gateway
ModelsOctober 29, 2026

Mistral Large 4 gets updated weights and lower prices on Vercel AI Gateway

Mistral Large 4 on Vercel's AI Gateway now runs on updated weights and costs less for input, cached input and output tokens. Developers already using the model do not…