claude-fable-5-1Claude Fable 5.1 is Anthropic's flagship model, released September 2026, and on APIMODELS it costs $5 per 1M input tokens and $25 per 1M output tokens — exactly half Anthropic's $10 / $50 list price. Prompt caching is cheaper here too — $0.22 per 1M cached reads against Anthropic's $0.25, and $5.778 per 1M cache writes against their $12.50 (5-minute) and $20 (1-hour), with both TTLs billed at the same rate — so all four token tiers sit below the official price, not just input and output. What changed against Fable 5 is agentic endurance: it more than doubles Fable 5 on Terminal-Bench-Science 0.1 (52.6% vs 24.7%, with Opus 5 at 29.0%), leads on Terminal-Bench 4.0 (55.8% vs 42.0%), CursorBench 3.2.0 (73.4% vs 70.5%), AutomationBench (31.4% vs 17.1%), OSWorld 2.0 strict (41.7% vs 36.1%) and Humanity's Last Exam with tools (65.0% vs 63.8%), and moves GDPval-AA v2 from 1723 to 1853 Elo. Anthropic demonstrated 38-hour unattended agent runs on it. The context window is 1M tokens with up to 128K output, and input is multimodal (text + image). Call it through the Anthropic Messages API (/v1/messages) — the native path for Claude Code, the Anthropic SDK and Anthropic-compatible clients like Cursor — with native tool use and prompt caching. Request params match Claude Fable 5, so switching is a one-line model-name change.
View complete API reference with all parameters and examples.
Enable real-time streaming responses with Server-Sent Events.
{
"model": "claude-fable-5-1",
"stream": true,
"max_tokens": 1024,
"messages": [...]
}Enable Claude to use tools and call functions.
{
"model": "claude-fable-5-1",
"max_tokens": 1024,
"tools": [{
"name": "get_weather",
"description": "Get current weather for a location",
"input_schema": {
"type": "object",
"properties": {
"location": {"type": "string", "description": "City name"}
},
"required": ["location"]
}
}],
"tool_choice": {"type": "auto"},
"messages": [{"role": "user", "content": "What's the weather in Tokyo?"}]
}Analyze PDF documents by sending them as base64 encoded content.
{
"model": "claude-fable-5-1",
"max_tokens": 1024,
"messages": [{
"role": "user",
"content": [{
"type": "document",
"source": {
"type": "base64",
"media_type": "application/pdf",
"data": "<base64_encoded_pdf>"
}
}, {
"type": "text",
"text": "Summarize this document."
}]
}]
}Get structured JSON responses that match your schema.
{
"model": "claude-fable-5-1",
"max_tokens": 1024,
"output_format": {
"type": "json_schema",
"schema": {
"type": "object",
"properties": {
"name": {"type": "string"},
"age": {"type": "integer"}
},
"required": ["name", "age"]
}
},
"messages": [{"role": "user", "content": "Extract info: John is 30 years old"}]
}Enable Claude to search the web for up-to-date information.
{
"model": "claude-fable-5-1",
"max_tokens": 1024,
"tools": [{
"type": "web_search_20250305",
"name": "web_search",
"max_uses": 5
}],
"messages": [{"role": "user", "content": "What's the latest news about AI?"}]
}| Parameter | Type | Required | Description |
|---|---|---|---|
| model | string | Yes | Model identifier (e.g., claude-fable-5-1) |
| messages | array | Yes | Array of message objects with role and content |
| max_tokens | integer | Yes | Maximum tokens in the response (1 - 128000) |
| system | string | No | System prompt to set context |
| stream | boolean | No | Enable streaming responses (SSE) |
| temperature | number | No | Sampling temperature (0.0 - 1.0) |
| top_p | number | No | Nucleus sampling threshold (0.0 - 1.0) |
| top_k | integer | No | Top-k sampling (0 - infinity) |
| stop_sequences | array | No | Sequences that stop generation |
| tools | array | No | Function calling tools definition |
| tool_choice | object | No | Tool selection strategy (auto/any/tool) |
| thinking | object | No | Enable extended thinking mode |
| output_format | object | No | Structured output with JSON schema |
View complete API reference with streaming, thinking, and more.
Billing: Cost = (input_tokens * input_price + output_tokens * output_price) / 1,000,000
$5 / $25 in-out, $0.22 cached read, $5.778 cache write
Terminal-Bench-Science 52.6% vs 24.7%
Multimodal input (text + image)
Native /v1/messages, tool use + prompt caching
Claude Fable 5.1 is a Large Language Model API provided by Anthropic. Claude Fable 5.1 is Anthropic's flagship model, released September 2026, and on APIMODELS it costs $5 per 1M input tokens and $25 per 1M output tokens — exactly half Anthropic's $10 / $50 list price. Prompt caching is cheaper here too — $0.22 per 1M cached reads against Anthropic's $0.25, and $5.778 per 1M cache writes against their $12.50 (5-minute) and $20 (1-hour), with both TTLs billed at the same rate — so all four token tiers sit below the official price, not just input and output. What changed against Fable 5 is agentic endurance: it more than doubles Fable 5 on Terminal-Bench-Science 0.1 (52.6% vs 24.7%, with Opus 5 at 29.0%), leads on Terminal-Bench 4.0 (55.8% vs 42.0%), CursorBench 3.2.0 (73.4% vs 70.5%), AutomationBench (31.4% vs 17.1%), OSWorld 2.0 strict (41.7% vs 36.1%) and Humanity's Last Exam with tools (65.0% vs 63.8%), and moves GDPval-AA v2 from 1723 to 1853 Elo. Anthropic demonstrated 38-hour unattended agent runs on it. The context window is 1M tokens with up to 128K output, and input is multimodal (text + image). Call it through the Anthropic Messages API (/v1/messages) — the native path for Claude Code, the Anthropic SDK and Anthropic-compatible clients like Cursor — with native tool use and prompt caching. Request params match Claude Fable 5, so switching is a one-line model-name change. Through APIMODELS platform, you can access this model via a unified API with transparent pay-as-you-go pricing. Current pricing: Input: $5, Output: $25 per 1M tokens.
Build intelligent conversational systems to automatically answer user queries and improve service efficiency.
Automatically write articles, emails, ad copy, and other text content to boost productivity.
Assist with code writing, debugging, and code review to accelerate software development.
Understand and analyze unstructured data, extract key insights, and generate summary reports.
Claude Fable 5.1 is available through APIMODELS at: Input: $5, Output: $25 per 1M tokens. Billing is pay-as-you-go — you only pay for what you generate.
Sign up at APIMODELS, get your API key, and call our unified API endpoint. We provide detailed API documentation with code examples in cURL, Python, and Node.js.
APIMODELS offers the same Claude Fable 5.1 model through our aggregation platform. We provide a unified API interface so you do not need separate accounts for each provider - one API key to access all models.
The gap is in how long it can keep working, not in how it chats. Anthropic's published numbers: Terminal-Bench-Science 0.1 goes from Fable 5's 24.7% to 52.6% (Opus 5 sits at 29.0%) — more than double; Terminal-Bench 4.0 55.8% vs 42.0%; CursorBench 3.2.0 73.4% vs 70.5%; AutomationBench 31.4% vs 17.1%; OSWorld 2.0 strict 41.7% vs 36.1%; GDPval-AA v2 moves from 1723 to 1853 Elo. Anthropic demonstrated 38-hour unattended agent runs on it. The choice is easy: take 5.1 for long-horizon agents, cross-repo code work and multi-step research. Both cost the same here ($5 / $25), so there is no cost argument for staying on Fable 5 — only an “it already works, don't touch it” argument.
Yes, on the native path. It speaks the Anthropic Messages API (/v1/messages): point ANTHROPIC_BASE_URL at https://api.apimodels.app/v1, set the model to claude-fable-5-1, and use your apimodels API key (see /docs/claude-code). Anthropic-compatible clients like Cursor work through their "custom Anthropic model" setting. Tools use the Anthropic schema ({name, input_schema}), and request parameters are identical to Fable 5 — migrating is a one-line model-name change.
Cached reads are $0.22 per 1M tokens and cache writes are $5.778 per 1M — both below Anthropic’s own list ($0.25 for reads; $12.50 and $20 for 5-minute and 1-hour writes). That puts all four tiers under the official price, not just input and output. Anthropic charges the two write TTLs differently; here both bill at the same $5.778, so passing ttl:"1h" costs you nothing extra. Usage is the standard cache_control:{"type":"ephemeral"} block — put the system prompt and the unchanging context first and keep it byte-identical between turns, or the prefix will not hit. Claude Code already uses caching by default; nothing to configure.
A 1M-token context window with up to 128K output tokens, and multimodal input (text + image). The full 1M context bills at the same input rate — there is no long-context surcharge. Worth knowing about the output ceiling: filling all 128K in one call costs about $3.2 (128K x $25/1M), so for long generations use stream:true — you see output as it arrives and you avoid a very long non-streaming request running into gateway timeouts.
By task shape. Fable 5.1 is the strongest of the three for long-horizon agents, terminal work and multi-step research, and it leads Opus 5 on the Terminal-Bench series, AutomationBench and OSWorld. Opus 4.8 is far cheaper per token ($3 / $13) and is the sensible default for ordinary conversation, single-shot code edits and structured output. All three share one API key and one endpoint, so routing per task is a change to the model field and nothing else.
On APIMODELS, Claude Fable 5.1 runs alongside 60+ models on one API key and one balance, so choosing is about fit, not lock-in. It supports 50% Below Official, Agentic Coding, 1M Context, Multimodal, and you can weigh it on price and capability against other Large Language Model models, then switch by changing a single model-name string — no new account or integration. Browse every Large Language Model option with live pricing at apimodels.app/models.
Claude Fable 5.1 supports: 50% Below Official, Agentic Coding, 1M Context, Multimodal. See the APIMODELS docs for full parameters and call examples.
Yes. APIMODELS exposes Claude Fable 5.1 through a single unified API and one key — no separate provider accounts, and no need to handle each provider's regional network access yourself.
We support Stripe (Visa, Mastercard, and other international cards) and Alipay. Credits are available instantly after payment.
How to get access, regional availability, and how this model compares with its alternatives.