How to Use Claude Fable Latest in Gemini CLI
Use Claude Fable Latest for difficult coding, research, and analysis that needs several rounds of checking. Anthropic currently identifies Fable 5.2 as its latest release, with support for large software migrations and mathematical proof formalization.
Prompts to try with Claude Fable Latest in Gemini CLI
Copy an example into Gemini CLI, or ask in your own words.
Use Claude Fable Latest at high reasoning effort to investigate the failing tests in this project. Reproduce the failure, trace its cause across components, implement a fix, and run the relevant checks before reporting.
Use Claude Fable Latest to plan a Python-to-Rust migration for this project's worker. Identify behavior that must stay identical, propose an incremental sequence, and implement the first independently testable step.
Use Claude Fable Latest to review the service boundaries in this repository. Trace dependencies, identify duplicated responsibilities, and write a concrete architecture proposal tied to the code.
Use Claude Fable Latest at high reasoning effort to evaluate this project's machine learning experiments. Compare the baseline and recent runs, propose an ablation, execute it if the environment permits, and report reproducibility limits.
Use Claude Fable Latest to complete the unfinished proof in this project's Lean files. State any additional assumptions, run the proof checker, and continue through ordinary errors until it passes or you identify a specific blocker.
How to set up Claude Fable Latest in Gemini CLI
Pick the way that suits you. Each takes about two minutes.
Use Spiral's MCP server
- Add the server
Or add it to ~/.gemini/settings.json yourself:
gemini mcp add --scope user --transport http spiral https://api.spiral.ai/mcp
- Sign in
Start gemini and run:
/mcp auth spiral
- Check it
In Gemini CLI, run /mcp. From your shell:
gemini mcp list
Use the Spiral skill
- Send Gemini CLI this message
Your agent reads Spiral's skill and sets itself up.
set up https://spiral.ai/SKILL.md
- Sign in or paste a key
Approve Spiral in the browser, or give the agent a key from the console.
Use the Spiral CLI
- Install and sign in
Installs the Spiral CLI, signs you in and adds its skill to your coding agents.
npx -y spiral-cli@latest init
Call the Spiral API
- Create an API key
In the console, under API keys.
- Run Claude Fable Latest
One request; the result comes back by polling or a webhook.
curl https://api.spiral.ai/v1/runs \ -H "Authorization: Bearer $SPIRAL_API_KEY" \ -H "Content-Type: application/json" \ -H "Idempotency-Key: $(uuidgen)" \ -d '{ "provider": "anthropic", "model": "~anthropic/claude-fable-latest", "input": { "messages": [ { "role": "system" } ] } }'
Then paste one of the prompts above: Gemini CLI finds Claude Fable Latest on Spiral, shows the price and runs it.
Claude Fable Latest providers and prices
Spiral runs Claude Fable Latest through Anthropic. Open a card for today's price, with every fee included.
Up to $0.21372 per run at the default inputs. $10.40 per 1M input tokens, $0.26 per 1M cached input tokens, $13.00 per 1M cache write tokens, $20.80 per 1M 1-hour cache write tokens, $52.00 per 1M output tokens.
Settings you can ask for
Ask Gemini CLI for any of these in your prompt.
| Setting | Allowed values | Default | Description |
|---|---|---|---|
| Max tokens | 1 to 128000 | The most output tokens to generate. On a reasoning model the reasoning tokens count toward it: a cap too low for the effort can end with no answer (finish_reason length). | |
| Reasoning effort | low, medium, high, xhigh, or max | high | How much the model reasons. |
| Parallel tool calls | on or off |
Every API field of Claude Fable Latest is on its tool page: open its card above.
What can you do with Claude Fable Latest in Gemini CLI?
Turn a difficult repository change into an implementation with checks using Claude Fable. Give the agent the relevant code, expected behavior, and completion criteria so it can carry the work through.
Plan and implement a software migration
Map existing behavior, migrate one component at a time, and compare results against the original implementation.
Investigate a persistent software bug
Trace a failure across components, test competing explanations, and produce a fix with evidence that it addresses the cause.
Review architecture across repositories
Compare dependencies and responsibilities across services, then turn the findings into a concrete change proposal.
Produce a technical research brief
Work from supplied sources to develop an analysis, resolve conflicting evidence, and finish a document with clear conclusions.
Evaluate machine learning experiments
Compare a baseline with proposed changes, plan ablations, and check whether measured improvements survive repeat runs.
Develop a checkable mathematical proof
Break a proof into lemmas, state assumptions, and work toward a formalization that a proof checker can validate.
More on Claude Fable Latest and Gemini CLI
Claude Fable Latest in other agents
Questions about Claude Fable Latest in Gemini CLI
Can Gemini CLI use Claude Fable Latest?
Yes. Add Spiral to Gemini CLI as an MCP server, and Gemini CLI can find and run Claude Fable Latest like any built-in tool.
How much does Claude Fable Latest cost in Gemini CLI?
Up to $0.21372 per run at the default inputs. $10.40 per 1M input tokens, $0.26 per 1M cached input tokens, $13.00 per 1M cache write tokens, $20.80 per 1M 1-hour cache write tokens, $52.00 per 1M output tokens. You pay from your Spiral balance as you go, with no subscription, and a failed run is not charged. Your Gemini CLI plan is separate.
Do I need an Anthropic account to use Claude Fable Latest in Gemini CLI?
No. Spiral runs Claude Fable Latest on its own Anthropic account. You sign in to Spiral once and never paste a key.
How long does it take to set up Claude Fable Latest in Gemini CLI?
About two minutes: add the server, then sign in, then check it.
Does Gemini CLI wait while Claude Fable Latest runs?
Claude Fable Latest runs in the background. Gemini CLI starts the run, checks on it and gives you the text, json when it's ready.
What is Claude Fable best suited for?
Difficult software changes, extended research, and analysis that needs repeated checking. Anthropic's current release, Fable 5.2, also supports work across large repositories.
How much context and output can Claude Fable handle?
The current Fable 5.2 release supports a one-million-token context window and up to 128,000 output tokens. Thinking uses part of the output allowance, so the maximum is not a promise of 128,000 tokens of visible text.
Which reasoning effort should I start with?
Start at high and evaluate the results on your actual tasks before increasing it. Leave enough output room for both thinking and the finished answer.
What weakness should I watch for on long tasks?
Fable 5.2 can end a turn while work remains. Give it a completion checklist and explicitly ask it to work through ordinary errors until the requested deliverable is finished.
Can Claude Fable help with mathematical proofs?
Yes. Anthropic reports that a team of Fable 5.2 agents formalized Fermat's Last Theorem; for your own work, ask for explicit assumptions and verification with a proof checker where available.
Start using Claude Fable Latest in Gemini CLI
One Spiral account for every tool your agent needs. No subscription; you pay per run.
Paste this into Gemini CLI:
set up https://spiral.ai/SKILL.md


