Claude Fable 5.1

Claude Fable 5.1 by Anthropic: $10 input and $50 output per 1M tokens, 1M context. Call it through the odnoga LLM gateway.

VisionToolsStreaming

Claude Fable 5.1 is served by Anthropic and called through odnoga with the same OpenAI-compatible request shape as every other model in the catalog. It accepts images alongside text. It supports tool and function calling. Its 1M-token context window sets how much input you can send in one request. Cached input is billed at $0.25 per 1M tokens, so repeated prefixes cost less.

Specification and price

VendorAnthropic
Model IDclaude-fable-5-1
Context window1M tokens
Max output128K tokens
Input / 1M tokens$10
Output / 1M tokens$50
Cached input / 1M tokens$0.25
CapabilitiesVision, Tools, Streaming

Prices are the vendor list price per million tokens as recorded in the odnoga catalog. Your odnoga cost is this price plus your plan margin — see pricing. Pricing

Call it through odnoga

const res = await fetch("https://api.odnoga.com/functions/v1/airouter-openai-compat/v1/chat/completions", {
  method: "POST",
  headers: {
    Authorization: `Bearer ${process.env.ODNOGA_API_KEY}`,
    "Content-Type": "application/json",
  },
  body: JSON.stringify({
    model: "claude-fable-5-1",
    messages: [{ role: "user", content: "Hello" }],
  }),
});

Same request shape for every vendor in the catalog — swap the model id and odnoga handles keys, routing, limits and cost accounting.

How to use Claude Fable 5.1

Reviewed 2026-09-14

Best for

  • Long-horizon agentic coding: multi-step refactors, migrations and tool-using coding agents that run for hours
  • Multistep research workflows that gather, cross-check and synthesize many sources
  • Demanding reasoning and knowledge work where Claude Opus 5 at higher effort still falls short
  • Document, spreadsheet and slide work — creating and editing complex office artefacts
  • Workloads with heavy prompt caching: cache reads are $0.25/MTok, 75% cheaper than Fable 5

Not the right pick when

  • Simple chat, classification or extraction — Claude Sonnet 5 or Haiku 4.5 are far cheaper for that
  • Most everyday workloads — Anthropic itself recommends starting with Claude Opus 5 first
  • Latency-sensitive interactive features: this is a frontier reasoning model, not a speed pick
  • Cost-sensitive high-volume traffic without caching or batch — at $10/$50 per MTok the bill grows fast

Practical tips through odnoga

  1. 01Pin claude-fable-5-1 in a managed prompt version so a future Fable upgrade never silently changes behaviour
  2. 02Turn on Anthropic prompt caching for repeated system prompts and tools — cache reads cost $0.25/MTok here
  3. 03Use the Batch API tier for offline evaluations: it supports batch and cuts cost on large test suites
  4. 04Run it in the evaluation lab against Opus 5 and Sonnet 5 on your own prompts before committing
  5. 05Set a per-tenant budget and a cheaper fallback model in odnoga before routing production traffic
  6. 06Use the effort parameter (low/medium/high/max/xhigh) to trade latency against reasoning depth per call

What a month costs

1,000 calls a month, 10,000 input tokens and 2,000 output tokens each, at this model’s catalog price:

Input (10M tokens)$100
Output (2M tokens)$100
Vendor cost per month$200

Vendor list price only — your odnoga price adds your plan margin, and cached input or answer reuse lowers it. odnoga records both numbers per request.

Claude Fable 5.1 compared

ModelContextInput / 1MOutput / 1MCapabilities
Claude Fable 5.11M$10$50Vision, Tools, Streaming
Claude 3.5 Sonnet200K$3$15Vision, Tools, JSON mode, Streaming
Claude Fable 51M$10$50Vision, Tools, Streaming

Questions

What does Claude Fable 5.1 cost?
$10 per million input tokens and $50 per million output tokens — the same as Fable 5. Cache reads dropped 75% to $0.25 per million tokens; cache writes are $12.50 (5-minute) and $20 (1-hour).
When should I pick Fable 5.1 over Claude Opus 5?
Anthropic recommends starting with Opus 5 for most workloads. Pick Fable 5.1 for demanding reasoning and long-horizon agentic work, or when your evals on Opus 5 at higher effort still fall short.
What is the difference between Fable 5.1 and Mythos 5.1?
They are the same underlying model with different safeguard levels. Mythos 5.1 is in limited availability; Fable 5.1 is the generally available endpoint.
Can I switch from Fable 5 to Fable 5.1 without code changes?
Yes. Through odnoga you change the pinned model in the managed prompt version or pass claude-fable-5-1 as the model parameter — the API shape is unchanged.

All models · Pricing · Docs · Compare models in the evaluation lab

Sources: Anthropic — Introducing Claude Fable 5.1 and Mythos 5.1, Claude Platform Docs — Fable 5.1 overview, Claude Platform Docs — Pricing

One gateway, every model.