Claude Fable 5.1 is designed for demanding reasoning and long-horizon agentic work, especially when Claude Opus 5 at higher effort is insufficient. It provides a 1M-token context window and up to 128K output tokens, accepts text and image input, and uses adaptive thinking that is always on. It improves long-running agentic coding, multistep research, document, spreadsheet, and slide workflows. Use model ID anthropic/claude-fable-5-1 through HaoAI's OpenAI-compatible or Anthropic-compatible API.
1M context window
128K max output tokens
Released: 2026-09-01
Protocols:OpenAIAnthropic
Capabilities:VisionFunction CallingReasoningPrompt CachingPDF InputWeb Search
Capabilities
- Long-horizon reasoningDesigned for demanding reasoning and agent workflows that operate across many steps and long contexts.
- Agentic coding and researchImproved performance for long-running coding agents, multistep research, and complex tool-driven work.
- Documents and office workflowsSupports complex document, spreadsheet, and presentation-oriented workflows.
- Adaptive thinkingAdaptive thinking is always enabled and reasoning depth is controlled through the effort setting.
- Multimodal inputAccepts text and image input and produces text output.
Pricing
PriceRate 0.25x
HaoAI
Official
Input
$2.5/M
$10/M
Output
$12.5/M
$50/M
Cache read
$0.063/M
$0.25/M
Cache creation
$3.13/M
$12.5/M
Cache creation 5m
$3.13/M
$12.5/M
Cache creation 1h
$5/M
$20/M
Code Examples
import anthropic
client = anthropic.Anthropic(
base_url="https://api.hao.ai/anthropic",
api_key="YOUR_HAOAI_API_KEY",
)
message = client.messages.create(
model="anthropic/claude-fable-5-1",
max_tokens=1024,
messages=[
{"role": "user", "content": "Hello!"}
],
)
print(message.content[0].text)Related Models
Frequently asked questions
HaoAI bills based on actual request usage. Input, output, cache, and any other applicable rates follow the live pricing table on this page. Final request costs are available in the console.