A.I.
Frontier models
Claude Fable 5.1
Claude Fable 5.1 is Anthropic’s most capable widely released model, aimed at long-horizon agents, research, and document, spreadsheet, and slide work. The API id is claude-fable-5-1. It ships on the Claude API, Amazon Bedrock, Google Vertex AI, and Microsoft Foundry.
Mythos 5.1 (claude-mythos-5-1) matches Fable 5.1 capabilities and pricing shape but is available to Project Glasswing participants only.
- Status
- Current
- AA score
- 53 (max effort)Artificial Analysis Intelligence Index, one independent score across many tests, fetched 22 Sep 2026
- Price
- $10 in, $50 outUS dollars per million tokens (word pieces) in and out, unless noted
Training
Anthropic lists a reliable knowledge and training data cutoff of June 2026. Parameter count, training FLOPs, and the full post-training recipe for this SKU are not published on the models overview or Fable pages.
Inference
Adaptive thinking is always on. Default effort is high. Relative to the rest of the current Claude lineup, Fable is the slower latency class. Effort steers thinking depth. Per-message effort mid-conversation is available as a beta on the Claude API.
Context
The documented context window is 1 million tokens. Max output is 128K tokens. Inputs are text and images; output is text. The model supports multilingual use, vision, and tool use.
Methods
Anthropic names an effort parameter, tools, and an agentic coding and research focus for this SKU. Forced tool use (tool_choice type any or a named tool) is not supported on Fable 5.1.
Benchmarking
Numbers below are from Anthropic’s Fable 5.1 announce, with production safeguards on for Fable. Anthropic’s GPT-5.6 Sol column is Anthropic’s harness, not OpenAI’s own Sol scores.
| Eval | Fable 5.1 | Fable 5 | Opus 5 | GPT-5.6 Sol (Anthropic column) |
|---|---|---|---|---|
| Terminal-Bench-Science 0.1 | 52.6% | 24.7% | 29.0% | 22.4% |
| Terminal-Bench 4.0 | 55.8% (Mythos 5.1 60.9%) | 42.0% | 52.3% | 37.3% |
| GDPval-AA v2 (score) | 1853 | 1723 | 1824 | 1711 |
| OSWorld 2.0 partial | 77.9% | 72.9% | 75.4% | - |
| OSWorld 2.0 strict | 41.7% | 36.1% | 39.6% | - |
| Humanity's Last Exam (no tools) | 60.9% | 57.8% | 56.6% | - |
| Humanity's Last Exam (with tools) | 65.0% | 63.8% | 63.6% | - |
| AutomationBench | 31.4% | 17.1% | 26.9% | 19.6% |
| CursorBench 3.2.0 | 73.4% | 70.5% | 70.0% | 67.2% |
Anthropic notes a Terminal-Bench-Science standard error of about ±3.5 to 4.5 points. OSWorld 2.0 here is the August 2026 task release and is not comparable to older OSWorld 2.0 prints. Safeguard interventions scored as zeros on some OSWorld and AutomationBench cases for Fable. Cyber tasks that hit Fable safeguards were completed by Opus 4.8; biology by Opus 5.
Hardware
Anthropic trains and runs Claude on AWS Trainium, Google TPUs, and NVIDIA GPUs. Amazon is the primary cloud and training partner (Project Rainier). Anthropic reported more than 1 million Trainium2 chips in use as of April 2026. Which chip serves which Fable request is not published. Buyers purchase API tokens, not a named Claude GPU SKU. Weights are closed; self-host is not offered.
Access and pricing
Pricing is $10 input and $50 output per million tokens. Cache reads are 2.5% of base input (versus 10% on other current Claudes).
Mythos 5.1 (claude-mythos-5-1) matches Fable 5.1 capabilities and pricing shape but is available to Project Glasswing participants only.