How do I run Claude Fable 5.1 in Claude Code without a Claude subscription?
Point Claude Code at Standard Compute and pin anthropic/claude-fable-5.1: Claude Fable 5.1 is on every plan from $19 a month, drawing on a fixed compute budget where new requests get HTTP 402 at 100%. The same key works in OpenCode and Codex.
| Model | Anthropic's top generally available model, positioned by Anthropic for demanding reasoning and long-horizon agentic work. |
|---|---|
| Model ID | anthropic/claude-fable-5.1Short form claude-fable-5.1 also works |
| Base URL | https://api.stdcmpt.com/v1Claude Code uses https://api.stdcmpt.com |
| Context | 1M tokens at Anthropic, with 128K max outputA request here takes up to about 850K input and 50K output tokens |
| Input | Text and images |
| Reasoning | low, medium, high, xhigh or max, set per model on the Models page |
Copy a working config
Pick your agent. Each config pins Fable 5.1 for the main conversation. Select it next to Opus 5.5 on the Models page and keep the model on standardcompute if you want routing to decide when Fable is worth it.
# Install the Standard Compute launcher once
curl -fsSL https://standardcompute.com/cli/install.sh | sh
# Start Claude Code with anthropic/claude-fable-5.1 as the main model
standardcompute launch claude -- --model anthropic/claude-fable-5.1
# Or configure it yourself. No /v1 here: Claude Code adds /v1/messages.
export ANTHROPIC_BASE_URL=https://api.stdcmpt.com
export ANTHROPIC_AUTH_TOKEN=YOUR_API_KEY
export ANTHROPIC_MODEL=anthropic/claude-fable-5.1
# Set every slot: Claude Code's own claude-*-5-5 names return model_not_found
export ANTHROPIC_DEFAULT_OPUS_MODEL=anthropic/claude-fable-5.1
export ANTHROPIC_DEFAULT_SONNET_MODEL=anthropic/claude-fable-5.1
export ANTHROPIC_DEFAULT_HAIKU_MODEL=anthropic/claude-fable-5.1
claude
Replace YOUR_API_KEY with a key from the dashboard. This config reads your key from STANDARDCOMPUTE_API_KEY; export it with a key from the dashboard. Full setup: Claude Code, OpenCode, Codex CLI.
What it costs
| Per 1M tokens | Fable 5.1 | Fable 5 |
|---|---|---|
| Input | $10.00 | $10.00 |
| Cache hit | $0.25 | $1.00 |
| 5-minute cache write | $12.50 | $12.50 |
| Output | $50.00 | $50.00 |
Anthropic standard API rates. The whole context window bills at the same rate. Source: Anthropic API pricing, checked October 13, 2026.
Average coding-agent turn on our gateway, Sep 13 to Oct 12, 2026: 87% of the prompt is cache hits and the reply is about 700 tokens. Anthropic list prices.
Cache hits are most of the prompt but only about a ninth of this bill, so fresh input and output decide what a Fable session costs. The $20 Starter budget covers roughly 120 turns like this one.
What works differently here
- Claude Code's thinking toggle does nothing here. The Messages endpoint ignores the thinking parameter and returns no thinking blocks; the reasoning level comes from the default you set under Customize on the Models page.
- With the launcher, only the main conversation is pinned. Claude Code's subagents and background calls still ask for its built-in sonnet and haiku names, which Standard Compute treats as smart routing across the models you selected.
- A Standard Compute plan is not a Claude Pro or Max subscription. Claude Mythos 5.1, which Anthropic sells at the same rates with limited availability, is not served.
Other ways to run Claude Fable 5.1
| Anthropic APIPay per token | First-party. Batch requests are half price and prompt caching is yours to set up. |
|---|---|
| Claude Pro / MaxSubscription | Claude Code's native sign-in, with session and weekly limits instead of a token bill. |
| Bedrock / Vertex / FoundryPay per token | Cloud billing for procurement teams. On Bedrock the ID is anthropic.claude-fable-5-1. |
| Standard ComputeMonthly plan | One key for Claude Code, Codex and OpenCode, with Claude next to GPT, Gemini and open models. |
FAQ
Why does claude-fable-5-1 return model_not_found?
claude-fable-5-1 return model_not_found?That is Anthropic's own ID, with hyphens. Standard Compute uses the dotted form, anthropic/claude-fable-5.1, so IDs match across every vendor in the catalog.
Fable 5.1 or Opus 5.5?
Anthropic suggests Opus 5.5 for most work and Fable 5.1 when Opus at higher effort still falls short. Opus 5.5 lists at $4 / $20 per million tokens, so it stretches a budget further.
Fable 5.1 or Fable 5?
Fable 5.1. It costs no more on any line of the table above, and Anthropic now lists Fable 5 as a legacy model.
Do I need a Claude subscription to use Claude Code this way?
No. Claude Code only needs a base URL and a token that speak the Messages API. Your Claude login stays as it is, and running claude without the launcher still uses it.
Create your key
Sign in, create a key in the dashboard and paste it into the config above.