Can Claude Fable 5 handle million-token documents through the API?

Yes. Claude Fable 5 supports a 1,000,000-token context window.

When should I use Claude Fable 5 instead of Claude Opus 4.8?

Use Claude Fable 5 for advanced reasoning, large-context analysis, and agent workflows.

How does Claude Fable 5 compare with Claude Mythos 5?

Fable 5 includes safety classifiers while Mythos 5 has broader capabilities and restricted access.

Is Claude Fable 5 suitable for autonomous coding agents?

Yes. It is optimized for long-running software engineering and agentic workflows.

Does Claude Fable 5 support image inputs?

Yes. It supports vision-based reasoning with screenshots, diagrams, and visual documents.

What are the main limitations of the Claude Fable 5 API?

Some cybersecurity, biology, chemistry, and distillation requests are restricted.

Can Claude Fable 5 be used for scientific research workflows?

Yes. It performs strongly on research, analysis, and scientific reasoning tasks.

Why is Claude Fable 5 considered a Mythos-class model?

It belongs to Anthropic's newest frontier architecture focused on advanced reasoning and long-horizon task execution.

Affordable Claude Fable 5 API | text-to-text

Technical Specifications of Claude Fable 5

Item	Specification
Model Name	Claude Fable 5
Provider	Anthropic
API Model ID	claude-fable-5
Model Class	Mythos-class (public safeguarded version)
Input Types	Text, Images
Output Types	Text
Context Window	1,000,000 tokens
Maximum Output Tokens	Up to 128,000 tokens
Core Strengths	Long-horizon reasoning, agentic workflows, coding, scientific research, vision
Availability	Anthropic API, Claude apps, Amazon Bedrock, major cloud partners
Safety Architecture	Automatic fallback to Claude Opus 4.8 for certain high-risk domains
Companion Model	Claude Mythos 5 (restricted-access variant)

Technical specifications are based on Anthropic's official model documentation and launch materials.

What Is Claude Fable 5?

Claude Fable 5 is Anthropic's most capable generally available AI model and the first public release built on its new Mythos-class architecture. It is designed for demanding reasoning tasks, autonomous multi-step workflows, large-scale software engineering, scientific research assistance, and extended agentic operations.

Fable 5 shares its underlying architecture with the restricted-access Claude Mythos 5, but includes additional safety classifiers. If requests enter certain high-risk domains—such as offensive cybersecurity, hazardous biology, chemistry, or model distillation—the system automatically routes the interaction to the safer Claude Opus 4.8 model. Anthropic states these safeguards activate in fewer than 5% of average sessions.

Main Features of Claude Fable 5

Massive 1M-token context window, enabling analysis of entire codebases, books, or large enterprise document collections.
Long-horizon agentic reasoning, allowing the model to sustain complex multi-step tasks over extended interactions.
State-of-the-art software engineering, including repository-scale code generation, debugging, and refactoring.
Advanced multimodal understanding, with strong performance on image and visual reasoning tasks.
Scientific and knowledge-work optimization, making it suitable for research, legal analysis, finance, and technical documentation.
Built-in safety routing, where sensitive requests are transparently handled by Claude Opus 4.8 to reduce misuse risk.

Benchmark Performance of Claude Fable 5

Anthropic describes Claude Fable 5 as "state of the art on nearly all tested benchmarks" and reports that its advantage becomes more pronounced as tasks grow longer and more complex. Public launch materials highlight exceptional performance in:

Software engineering and large codebase reasoning.
Autonomous knowledge work and document analysis.
Scientific research and technical problem solving.
Vision and multimodal understanding.
Long-running agent workflows requiring sustained planning.

While Anthropic has not publicly released a full benchmark table for every evaluation suite, it positions Fable 5 above previous Claude generations, including Claude Opus 4.8, particularly for long-duration autonomous tasks.

Claude Fable 5 vs Claude Opus 4.8 vs Claude Mythos 5

Feature	Claude Fable 5	Claude Opus 4.8	Claude Mythos 5
Public Availability	Yes	Yes	Limited access
Context Window	1M tokens	Large context	1M tokens
Long-Horizon Agent Work	Excellent	Strong	Excellent
Software Engineering	State-of-the-art	Advanced	State-of-the-art
Safety Guardrails	Yes	Standard	Reduced
High-Risk Cyber/Bio Tasks	Falls back to Opus 4.8	Supported within standard policies	Greater capability for trusted partners
Primary Audience	General developers & enterprises	General-purpose AI users	Approved cybersecurity and research organizations

Limitations

Some cybersecurity, biology, chemistry, and model-distillation requests are automatically redirected to Claude Opus 4.8.
The conservative safeguard system can occasionally trigger on benign requests.
Access to the unrestricted Mythos-level capabilities remains limited to trusted organizations through Project Glasswing.

Typical Use Cases

Repository-scale software development and code migration.
Enterprise document analysis and knowledge management.
Long-form research and technical writing.
Scientific literature review and hypothesis generation.
AI agents that require persistent multi-step planning.
Legal, financial, and consulting workflows involving very large datasets.

Pricing for Claude Fable 5

Explore competitive pricing for Claude Fable 5, designed to fit various budgets and usage needs. Our flexible plans ensure you only pay for what you use, making it easy to scale as your requirements grow. Discover how Claude Fable 5 can enhance your projects while keeping costs manageable.

Comet Price (USD / M Tokens)	Official Price (USD / M Tokens)	Discount
Input:$8/M Output:$40/M	Input:$10/M Output:$50/M	-20%

Sample code and API for Claude Fable 5

Access comprehensive sample code and API resources for Claude Fable 5 to streamline your integration process. Our detailed documentation provides step-by-step guidance, helping you leverage the full potential of Claude Fable 5 in your projects.

Python
JavaScript
Curl

import os

import anthropic

client = anthropic.Anthropic(
    base_url="https://api.cometapi.com",
    api_key=os.environ["COMETAPI_KEY"],
)

message = client.messages.create(
    model="claude-fable-5",
    max_tokens=4096,
    thinking={"type": "adaptive", "display": "summarized"},
    output_config={"effort": "medium"},
    messages=[
        {
            "role": "user",
            "content": "Analyze the trade-offs between microservices and monolithic architectures",
        }
    ],
)

for block in message.content:
    if block.type == "thinking" and block.thinking:
        print(f"[Thinking]
{block.thinking}
")
    elif block.type == "text":
        print(block.text)

Python Code Example

import os

import anthropic

client = anthropic.Anthropic(
    base_url="https://api.cometapi.com",
    api_key=os.environ["COMETAPI_KEY"],
)

message = client.messages.create(
    model="claude-fable-5",
    max_tokens=4096,
    thinking={"type": "adaptive", "display": "summarized"},
    output_config={"effort": "medium"},
    messages=[
        {
            "role": "user",
            "content": "Analyze the trade-offs between microservices and monolithic architectures",
        }
    ],
)

for block in message.content:
    if block.type == "thinking" and block.thinking:
        print(f"[Thinking]\n{block.thinking}\n")
    elif block.type == "text":
        print(block.text)

JavaScript Code Example

import Anthropic from "@anthropic-ai/sdk";

const anthropic = new Anthropic({
  apiKey: process.env.COMETAPI_KEY,
  baseURL: "https://api.cometapi.com",
});

const message = await anthropic.messages.create({
  model: "claude-fable-5",
  max_tokens: 4096,
  thinking: {
    type: "adaptive",
    display: "summarized",
  },
  output_config: {
    effort: "medium",
  },
  messages: [
    {
      role: "user",
      content: "Analyze the trade-offs between microservices and monolithic architectures",
    },
  ],
});

for (const block of message.content) {
  if (block.type === "thinking" && block.thinking) {
    console.log(`[Thinking]\n${block.thinking}\n`);
  } else if (block.type === "text") {
    console.log(block.text);
  }
}

Curl Code Example

# Get your CometAPI key from https://www.cometapi.com/console/token
# Export it as: export COMETAPI_KEY="your-key-here"
curl https://api.cometapi.com/v1/messages \
  --header "Authorization: $COMETAPI_KEY" \
  --header "content-type: application/json" \
  --data \
'{
  "model": "claude-fable-5",
  "max_tokens": 4096,
  "thinking": {
    "type": "adaptive",
    "display": "summarized"
  },
  "output_config": {
    "effort": "medium"
  },
  "messages": [
    {
      "role": "user",
      "content": "Analyze the trade-offs between microservices and monolithic architectures"
    }
  ]
}'

Versions of Claude Fable 5

The reason Claude Fable 5 has multiple snapshots may include potential factors such as variations in output after updates requiring older snapshots for consistency, providing developers a transition period for adaptation and migration, and different snapshots corresponding to global or regional endpoints to optimize user experience. For detailed differences between versions, please refer to the official documentation.

version
claude-fable-5