GPT Image 2.5 is live — OpenAI's newest image model, targeted edits that leave the rest of the frame alone

Claude Fable 5.1 — Anthropic's Most Capable Tier, 1M Context

Claude Fable 5.1 is Anthropic's most capable widely released model, built for the most demanding reasoning and long-horizon agentic work. Claude Fable 5.1 reads a 1,000,000-token context window, returns up to 128,000 tokens, and thinks on every request with the depth set by five effort rungs. On reAPI, current Claude Fable 5.1 token rates are shown in the pricing table below.

Highlights
Thinking always on · five effort rungs · long-horizon agentic work
Best for
Hard reasoning, agentic coding, research, long documents
Input
Chat messages, OpenAI-compatible format
Output
Text, streaming optional
Claude Fable 5.1

Ask Claude Fable 5.1

Start with the prompt in the composer or choose an example.

Enter to send · Shift+Enter for a new line

Pricing

Credit-based — 1 credit = $0.001 USD. Pay only for completed generations.

30% off official pricing· Input / output tokens

First test cost
$0.04242 credits

One 1K-token prompt with a 1K-token answer at the cheapest tier.

Testing budget guide
Add credits
$10
≈ 238 tests
$50
≈ 1190 tests
$100
≈ 2380 tests
Input
$7.00
Price · 1M tokens
Output
$35.00
Price · 1M tokens
Cache write
$8.75
Price · 1M tokens
Cache read
$0.70
Price · 1M tokens

What is Claude Fable 5.1?

Claude Fable 5.1 is Anthropic's most capable widely released model, positioned for the most demanding reasoning and long-horizon agentic work — the jobs that run for many turns and have to be right at the end of them. On reAPI, Claude Fable 5.1 is served through an OpenAI-compatible chat completions endpoint with a 1,000,000-token context window whose maximum is also its default, up to 128,000 output tokens, and thinking that is always on. Claude Fable 5.1 succeeds Claude Fable 5 in the same tier at the same per-token price. Billing is per token from your reAPI balance, with input and output priced separately; thinking tokens count as output.

What you can build with this model

Real-world workflows and production use cases you can build and ship with this model.

Claude Fable 5.1 driving a long multi-step task to completion

Agentic work that runs for hours, not turns

Anthropic positions Claude Fable 5.1 around long-horizon agentic work, and the practical consequence is that a single request can think for many minutes before it says anything. That is the shape of job to send it: port the module and keep going until the suite is green, work the incident from the first alert to the root cause, drive the migration through every step that can fail. Give Claude Fable 5.1 the whole task specification up front and let it run — the model is tuned for autonomy, and prompts written for earlier models are often too prescriptive and make its output worse.

Read the API docs
Claude Fable 5.1 reading an entire repository and its history in a single request

The whole system in one request

A 1,000,000-token context window means the service, its tests and its recent history go in as one request, so Claude Fable 5.1 reasons about the actual system rather than a summary someone wrote of it. The window's maximum is also its default, so nothing has to be requested to get it. Claude Fable 5.1 shares its tokenizer with Claude Opus 4.8, which means a cost model built on 4.7 or 4.8 token counts carries over unchanged and only the per-token rate moves.

Claude Fable 5.1 weighing a high-stakes decision with the full evidence in context

The calls that have to be right

There is a class of request where a wrong answer costs more than a month of inference: a contract review, an architectural commitment, a security finding, a research synthesis across hundreds of pages. Claude Fable 5.1 is the tier to send those to, at max effort if the decision warrants it. Route the routine turns to a cheaper model and keep Claude Fable 5.1 for the work you would want your most senior colleague to do — the escalation is a model string, not a second integration.

Why reAPI for Claude Fable 5.1

Pay-as-you-go pricing

Anthropic publishes per-million-token rates for Claude Fable 5.1 input and output. No subscription, no minimum commitment, pay-as-you-go from your reAPI balance. The live numbers are in the pricing table on this page.

OpenAI-compatible drop-in

Claude Fable 5.1 is served from reAPI's standard chat completions endpoint. If your code already speaks that shape, adopting Claude Fable 5.1 is a base URL, a key and a model string — the same SDKs work once the base URL points at reapi.ai. No vendor SDK to add, no bespoke client, and no second account to open just to reach Anthropic's top tier.

One key, one balance, every model

The reAPI key that calls Claude Fable 5.1 is the same key that generates images and video on this platform, drawing from one balance. Send the routine turns to a cheaper chat model and escalate to Claude Fable 5.1 by changing a model string. Mixing vendors costs nothing extra here: Claude Fable 5.1 and the flagship models from other labs sit behind the same endpoint.

Claude Fable 5.1 vs GPT-6 Astra

The two flagship tiers on this endpoint cost the same per token on reAPI, so price decides nothing and behaviour decides everything. Claude Fable 5.1 is the always-thinking, long-horizon agentic model with the full effort ladder and tool calling in place; GPT-6 Astra is the flagship of the other family, with a slightly larger window and a different set of restrictions. The rows below are the ones that actually pick one for a given route.

Capability
Claude Fable 5.1 on reAPI
GPT-6 Astra
Vendor positioning
Anthropic's most capable widely released model, for the most demanding reasoning and long-horizon agentic work
OpenAI's most capable model, built for the hardest end-to-end work
Context and output limits
1,000,000-token context, the maximum is also the default; 128,000-token maximum output
1,050,000-token context; 128,000-token maximum output
Reasoning effort
All five rungs reach the model: low, medium, high, xhigh, max. Thinking cannot be disabled
Five rungs reachable here — low, medium, high, xhigh, max. No default stated by the vendor
Tool calling on this endpoint
Tools pass through; only a forced tool choice is rejected
Not available — the vendor requires its separate Responses API for tools on this model
Price shape
Current input and output rates are shown in the pricing table
Prompts past a length threshold re-price the whole request at a higher tier
Operating constraints
Requires 30-day data retention; refusals arrive as finish_reason content_filter on a successful response
No retention floor of that kind; sampling parameters are rejected outright

Comparison reflects behavior documented by each vendor at the time of writing, plus the request surface each model exposes through this endpoint. Rates are described as relationships rather than amounts; the live numbers for Claude Fable 5.1 are in the pricing table above.

Ship Claude Fable 5.1 in three steps

  1. step 01

    Create a reAPI API key

    Sign in to reAPI and create a key under API keys. That one key reaches Claude Fable 5.1 and every other model on the platform, billed from a single balance.

    Open
  2. step 02

    Point your client at reapi.ai

    Set your base URL to reAPI's API and your key to the one you just created. Any client that already speaks OpenAI Chat Completions works unchanged — no SDK swap is needed to reach Claude Fable 5.1.

    Open
  3. step 03

    Send claude-fable-5-1 as the model

    Set the model field to claude-fable-5-1 and post your messages. Turn on streaming, leave room in the token budget for thinking, and pick the effort deliberately: Claude Fable 5.1 thinks on every request, and effort is what decides how long for.

    Open
docs/api/claude-fable-5-1

API reference

Drop-in code and the full parameter table.

View full reference
curl --no-buffer --request POST \
  --url https://reapi.ai/api/v1/messages \
  --header "x-api-key: YOUR_API_KEY" \
  --header "Content-Type: application/json" \
  --data '{
  "model": "claude-fable-5-1",
  "messages": [
    {
      "role": "user",
      "content": "Explain this design clearly."
    }
  ],
  "max_tokens": 4096,
  "output_config": {
    "effort": "high"
  },
  "stream": true
}'

Parameter validation failed

Rejected synchronously with the failing field named. Check enum values, ranges, and URL fields — nothing is charged.

Authentication or balance issue

401 means a missing or invalid Bearer key; 402 means the reserve exceeds your balance. Manage keys and credits in the console.

Content or material rejected

Reference material and outputs pass automated moderation. Rejected tasks fail with a clear error and a full refund.

Task failed or timed out

A task that reaches a failed state is never charged — the reserve refunds automatically. Keep the task id and retry when ready.

Frequently asked questions

Common questions about this model.

Claude Fable 5.1 is billed pay-as-you-go per token from your reAPI balance, with separate input and output rates. Thinking tokens are billed as output. Current numbers are in the pricing table on this page.

Related models

Explore more models in the same category.

View all models
start building

Ready to ship?

Try it in the playground or grab an API key to integrate now.

Get API keyView API docs