Claude Fable 5: pricing, performance, and how to route requests

Route requests to
Claude Fable 5
with Merge Gateway

Apply your own routing policies, reduce token costs automatically, and see every routing decision in real time with Merge Gateway.

How Claude Fable 5 performs*

Intelligence - general reasoning and knowledge

Coding - code generation and problem-solving

*Performance data is provided by Artificial Analysis and is subject to change.

What Claude Fable 5 costs to run

| Vendor | Input / 1M tokens | Output / 1M tokens | Zero data retention | | --- | ---: | ---: | --- | | Anthropic | $10.00 | $50.00 | No |

Test Claude Fable 5
with Gateway’s Simulator

See a prompt's output, token spend, latency, and more with Claude Fable 5.

Route requests to Claude Fable 5 in minutes

To get started in seconds, add our Gateway Implementation skill to your project, or pick your preferred SDK below. Check out our other quick start skills here.

Install the Merge Gateway SDK

Python

1$ pip install merge-gateway-sdk

Send a request

Python

1from merge_gateway import MergeGateway
2
3client = MergeGateway(api_key="YOUR_API_KEY")
4
5response = client.responses.create(
6    model="openai/gpt-5.2",
7    input=[
8        {"type": "message", "role": "system", "content": "You are a helpful programming tutor. Explain the concepts clearly with practical examples."},
9        {"type": "message", "role": "user", "content": "Explain the concept of recursion in programming with a simple set of examples."},
10    ],
11)
12
13print(response.output[0].content[0].text)

Try a diffrent model

Swap the model string to route to a different provider. No other code changes needed.

Anthropic

1response = client.responses.create(
2    model="anthropic/claude-sonnet-4-20250514",
3    input=[
4        {"type": "message", "role": "system", "content": "You are a helpful programming tutor. Explain the concepts clearly with practical examples."},
5        {"type": "message", "role": "user", "content": "Explain the concept of recursion in programming with a simple set of examples."},
6    ],
7)

Point to Gateway

Python

1from openai import OpenAI
2
3client = OpenAI(
4    api_key="YOUR_API_KEY",
5    base_url="https://api-gateway.merge.dev/v1/openai",
6)

Send a request

Use the standard chat.completions.create method. No provider prefix needed on the model name.

Python

1response = client.chat.completions.create(
2    model="gpt-5.2",
3    messages=[
4        {"role": "system", "content": "You are a helpful programming tutor. Explain the concepts clearly with practical examples."},
5        {"role": "user", "content": "Explain the concept of recursion in programming with a simple set of examples."},
6    ],
7)
8
9print(response.choices[0].message.content)

Install packages

1npm install merge-gateway-ai-sdk-provider ai

Create the provider

TypeScript

1import { createMergeGateway } from "merge-gateway-ai-sdk-provider";
2
3const gateway = createMergeGateway({
4  apiKey: "YOUR_API_KEY",
5});

Send a request

Use generateText to send a request. Model names use the provider/model format.

TypeScript

1import { generateText } from "ai";
2
3const { text } = await generateText({
4  model: gateway("openai/gpt-4o"),
5  prompt: "Explain the concept of recursion in programming with a simple set of examples.",
6});
7
8console.log(text);

If you already have @ai-sdk/openai installed, point it at Gateway with a base URL change:

TypeScript

1import { createOpenAI } from "@ai-sdk/openai";
2
3const gateway = createOpenAI({
4  apiKey: "YOUR_API_KEY",
5  baseURL: "https://api-gateway.merge.dev/v1/ai-sdk",
6});
7
8// All generateText/streamText calls work unchanged

Install the Merge Gateway SDK

Anthropic SDK

1from anthropic import Anthropic
2
3client = Anthropic(
4    api_key="YOUR_API_KEY",
5    base_url="https://api-gateway.merge.dev/v1/anthropic",
6)
7
8message = client.messages.create(
9    model="claude-sonnet-4-20250514",
10    max_tokens=1024,
11    messages=[
12        {"role": "user", "content": "Explain the concept of recursion in programming with a simple set of examples."},
13    ],
14)
15
16print(message.content[0].text)

Explore other models available in Merge Gateway

Amazon Nova 2 Lite

Amazon Nova Lite

Amazon Nova Micro

Amazon Nova Pro

Bland Speech v3

Clairic

Claude 3.7 Sonnet

Claude 4 Opus 20250514

Claude Fable 5

Claude Fable 5.1

Claude Haiku 4.5 (20251001)

Claude Haiku 5.5

Claude Opus 4.1 (20250805)

Claude Opus 4 (20250514)

Claude Opus 4.5 (20251101)

Claude Opus 4.6

Claude Opus 4.7

Claude Opus 4.8

Claude Fable 5 FAQ

In case you have any other questions on Claude Fable 5, we've answered a few more below. It's worth noting that the information below was written in July, 2026 and is subject to change.

Heading

What other models does Anthropic offer?

Anthropic ships a tiered Claude family so teams can match capability to task complexity and budget. Here are some other models Anthropic supports:

Claude Opus 5: The current Opus-tier flagship, a same-price successor to Opus 4.8, built for complex agentic coding and enterprise work at roughly half of Fable 5's per-token cost

Claude Sonnet 5: Anthropic's default mid-tier model, matching Fable 5's 1M-token context window at a fraction of the price, built for teams that want near-frontier coding and reasoning without frontier-tier spend

Claude Opus 4.8: The prior-generation flagship that Opus 5 replaced, still available at its original rate and notable as the model Fable 5 automatically falls back to when a request trips its biology or chemistry safety classifier

How does Claude Fable 5 differ from Anthropic's other models?

Fable 5 sits at the top of Anthropic's price ladder, built on the same underlying model as the invitation-only Claude Mythos 5 but hardened with additional safety classifiers for general release.

Pricing: Fable 5 costs $10 per million input tokens and $50 per million output tokens, exactly double Claude Opus 5's $5 and $25, and well above Claude Sonnet 5's standing rate of $3 and $15

Benchmark scores: on the Artificial Analysis Intelligence Index, Fable 5 scores approximately 60, within a point of Opus 5's 61 and clearly ahead of Sonnet 5's 53, putting the two priciest Claude models effectively tied at the top of the index rather than Fable 5 holding a clean lead

Reasoning mode: Fable 5 runs adaptive thinking always-on, the only current Claude model where this can't be switched off, while Opus 5 and Sonnet 5 offer it as a toggle and Haiku 4.5 relies on classic extended thinking instead

Speed and latency: Anthropic rates Fable 5's comparative latency as "slower," and independent testing puts its time-to-first-token past a minute on complex prompts, well behind Opus 5's "moderate" rating and Sonnet 5's "fast" one

Unique capabilities: Fable 5 leads the lineup on vision-heavy tasks, including reconstructing source code from screenshots and reading exact values off scientific figures, and on sustained autonomous agent runs, and it's the only model carrying Mythos-class safety classifiers for cybersecurity and biology or chemistry queries

Fable 5 is the right call for vision-heavy analysis, long-running autonomous agents, and cybersecurity- or biology-adjacent reasoning where the extra safety tooling and headroom matter. For general agentic coding or knowledge work, Opus 5 lands within a point of the same index score at half the price.

What models should I consider using alongside Claude Fable 5?

No single model is optimal for every task. Here are models worth pairing with Claude Fable 5 depending on what your product needs:

GPT-5.4 Mini for high-volume classification, extraction, or routing tasks where Fable 5's reasoning depth and $50-per-million output price are unnecessary overhead

Gemini 3.1 Pro Preview when the job is retrieval or synthesis across large multi-document or multimodal context rather than the multi-step reasoning Fable 5 specializes in

Grok 4.5 for tasks that need current-events awareness or real-time data grounding, a gap Claude models don't specialize in

What are the challenges of using Claude Fable 5 in my product?

Like any production LLM, Claude Fable 5 comes with tradeoffs worth planning for:

Provider dependency: routing every request straight to Anthropic's API means an outage, a rate-limit change, or a future deprecation of the claude-fable-5 model ID can take your product down with it unless a fallback path is already built

Cost at scale: at $50 per million output tokens, the highest output price in Anthropic's current lineup, and with Fable 5's traces running notably verbose under always-on adaptive thinking, unmanaged usage compounds fast on agentic workflows that chain multiple calls

Marginal gains over cheaper siblings: Fable 5's Intelligence Index score sits within a point of Opus 5's, and its reliable knowledge cutoff (January 2026) is actually older than Opus 5's (May 2026), so the 2x price premium doesn't automatically buy a proportional jump in capability or freshness for every workload

Latency at default reasoning effort: time-to-first-token has been measured well past a minute on complex prompts, which rules Fable 5 out for synchronous, latency-sensitive UI paths without a faster fallback model in front of it

Access can be gated or paused: Fable 5 rolled out in stages contingent on capacity and was briefly pulled back after launch to comply with export-control requirements, a reminder that the newest, most-restricted tier can see availability change with little warning

Why should I use Merge Gateway to route LLM requests with Claude Fable 5 and every other model?

Using Claude Fable 5 through Merge Gateway gives you access to the model itself and the infrastructure layer around it:

One API, every provider: call Claude Fable 5 and every other major model through the same endpoint and API key, and move between providers by changing a model string instead of rewriting integration code

Intelligent routing and automatic failover: Merge detects Anthropic outages and reroutes traffic automatically, and cost-, latency-, or quality-aware routing policies typically trim 40-60% off spend without any changes to your application

Cost governance: cap Fable 5 spend with hard or soft project budgets, and see every request rolled up by model, project, and tag in one billing view that spans all your providers

Build Your Own Router: decide what "best" means for your traffic by weighting curated ML benchmarks or your own eval scores, then let the router score every candidate model per request and explain, in plain language, why it picked the winner

Security and compliance controls: screen requests with DLP rules and prompt injection defenses before they reach Anthropic, and lock in per-project model and region policies without building any of that logic yourself

How can I start routing requests to Claude Fable 5 via Merge Gateway?

Getting Claude Fable 5 running through Merge Gateway takes a few minutes:

1. Create an account and get your API key from the dashboard.

2. Install the Merge Gateway SDK: run pip install merge-gateway-sdk (Python) or npm install merge-gateway-sdk (Node). Alternatively, if you're already using the OpenAI SDK, set base_url = "https://api-gateway.merge.dev/v1/openai" and your existing code works as-is.

3. Make your first request using the provider/model format. For Claude Fable 5, the model string is anthropic/claude-fable-5. Swap the model string to route to any other provider without changing anything else.

4. Configure a routing policy in the dashboard to set failover behavior, cost limits, and optimization strategy. Your first policy can be as simple as naming Claude Fable 5 as primary with one fallback.

Full setup instructions and SDK references are in the Merge Gateway docs.