<MCP.Playground/>
online

Claude Fable 5.1

Runs MCP toolsReasoning

Anthropic’s top-tier model, released 1 September 2026. The biggest gains Anthropic reports are in long-running agent work, which is where MCP servers get exercised hardest.

Vendor
Anthropic
Released
1 September 2026
Context
1M tokens
Input
Text, image, file

Paste a server URL, pick a model, and watch it call your tools in a real conversation. No install.

No server of your own? Leave it blank and use one of the public mock servers.

In your browser

MCP Agent Studio

Server has a public URL? Chat with Claude Fable 5.1 on it, watch every tool call, and compare models side by side.

Open MCP Agent Studio →

On your computer

MCP Playground Desktop

Server runs on localhost or over stdio? The browser can’t reach it. The desktop app can, with Claude Fable 5.1 in Agent Studio. Download free

Mac (Apple Silicon)Mac (Intel)Windowsor Microsoft Store

MCP Evals

Can Claude Fable 5.1 actually complete tasks with your server?

One chat shows the tool calls work. An eval writes real tasks from your tool schemas, has Claude Fable 5.1 drive each one, and reports where it got the wrong answer even though every call succeeded. Pick Claude Fable 5.1 as the driver.

Run evals →See a sample

What Claude Fable 5.1 is

Claude Fable 5.1 succeeds Fable 5. Anthropic released it alongside Mythos 5.1, which is the same model with different safeguards and is limited to trusted-access programmes for cybersecurity and life-sciences work. Fable 5.1 is the generally available one. It has a 1M-token window, takes text, image and file input, and is on Anthropic’s API as claude-fable-5-1 as well as AWS, Google Cloud and Azure.

What Anthropic says

  • Terminal-Bench 4.0: 55.8%, against 42.0% for Fable 5.
  • AutomationBench: 31.4%, against 17.1% for Fable 5.
  • OSWorld 2.0: 77.9% partial pass and 41.7% strict pass, against 72.9% and 36.1% for Fable 5.
  • GDPval-AA v2: 1853, against 1723 for Fable 5.

These are the vendor’s own claims, not measurements of ours. Run the model against your own server to find out whether they hold for your tools.

How it reaches MCP

  • Natively. Anthropic’s Messages API can connect to remote MCP servers itself. On the API the model id is claude-fable-5-1.
  • Through Claude Code, Claude Desktop and claude.ai, all of which act as MCP clients
  • MCP Playground’s Agent Studio: paste your server URL above and run Fable 5.1 against it in the browser, with no Anthropic key

What to watch for

  • Its default effort differs by surface: High in Claude Code, Medium on claude.ai and in Claude Cowork. A server that works in Claude Code and struggles on claude.ai may be seeing less reasoning, not a different model.
  • Anthropic says Opus 5.5, released three weeks later, matches Fable 5.1 on most work and reports it ahead on Terminal-Bench 4.0 (66.4% against 55.8%). Run both on your tools before settling on Fable.
  • OpenRouter lists neither temperature nor seed for Fable 5.1, while Opus 5.5 takes temperature. Tool calls will vary from run to run. Run a prompt more than once before you conclude anything from one result.

Run the error-simulation mock and give Fable 5.1 a task that hits simulate_partial_success, then the same task on Opus 5.5. The difference to look for is whether each one reports the partial result or retries until it can claim success. All the mock servers →

Available here

ModelModel ID
Claude Fable 5.1anthropic/claude-fable-5.1

Go deeper

Frequently asked questions

What is the difference between Claude Fable 5.1 and Mythos 5.1?
Anthropic says they are the same model with different safeguards. Fable 5.1 is generally available. Mythos 5.1 is restricted to trusted-access programmes for cybersecurity and life-sciences professionals.
Does Claude Fable 5.1 support MCP?
Yes, natively. Anthropic’s API can connect to remote MCP servers directly, and Claude Code, Claude Desktop and claude.ai are all MCP clients. You can also run it against any server here without an Anthropic key.
Fable 5.1 or Opus 5.5 for agent work?
Anthropic says Opus 5.5 matches Fable 5.1 on most work, generates output faster, and scores higher on Terminal-Bench 4.0. Run both on the same prompt against your server. If the tool calls match, Opus 5.5 is the lighter choice.
What changed from Claude Fable 5?
Anthropic’s biggest reported gains are in agent work. Terminal-Bench 4.0 rose from 42.0% to 55.8% and AutomationBench from 17.1% to 31.4%. The window stays at 1M tokens.

Sources