How to Use Inkling Small (batch) in Muse Code
Inkling Small suits coding, rubric-based grading, and synthetic training examples. In the maker's evaluations, it beats Inkling on reasoning and agentic coding, while Inkling retains an advantage in factuality.
Prompts to try with Inkling Small (batch) in Muse Code
Copy an example into Muse Code, or ask in your own words.
Use Inkling Small to investigate the failing tests in this project. Reproduce one failure, trace its cause, make a focused fix, and run the relevant tests.
Use Inkling Small to compare this project's implementation with its documented requirements. List mismatches with file references and propose a test for each.
Use Inkling Small to review the current diff for correctness. Focus on changed behavior, boundary cases, and missing tests; support each finding with a code reference.
Use Inkling Small to draft synthetic examples for the classifier in this repository. Follow the existing label definitions, include ambiguous cases, and mark examples that need review.
Use Inkling Small with more thinking effort to investigate this project's intermittent test failure. Compare plausible causes against the code and test output, then propose the smallest experiment that distinguishes them.
How to set up Inkling Small (batch) in Muse Code
Pick the way that suits you. Each takes about two minutes.
Use Spiral's MCP server
- Install Muse Code
curl -fsSL https://dev.meta.ai/install.sh | bash
- Ask Muse to add the server
Start muse and ask:
Add the Spiral MCP server at https://api.spiral.ai/mcp
- Sign in
In your terminal, run:
muse mcp login spiral
Use the Spiral skill
- Send Muse Code 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 Inkling Small (batch)
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": "openrouter", "model": "thinkingmachines/inkling-small:batch", "input": { "messages": [ { "role": "system" } ] } }'
Then paste one of the prompts above: Muse Code finds Inkling Small (batch) on Spiral, shows the price and runs it.
Inkling Small (batch) providers and prices
Spiral runs Inkling Small (batch) through OpenRouter. Open a card for today's price, with every fee included.
Up to $0.005129 per run at the default inputs. $0.468 per 1M input tokens, $1.248 per 1M output tokens.
Settings you can ask for
Ask Muse Code for any of these in your prompt.
| Setting | Allowed values | Default | Description |
|---|---|---|---|
| Seed | a whole number | ||
| Min p | 0 to 1 | ||
| Top k | a whole number from 0 | ||
| Top p | 0 to 1 | ||
| Max tokens | 1 to 262144 | 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). | |
| Temperature | 0 to 2 | ||
| Presence penalty | -2 to 2 | ||
| Reasoning effort | none, minimal, low, medium, high, or max | high | How much the model reasons. |
| Frequency penalty | -2 to 2 | ||
| Repetition penalty | 0 to 2 |
Every API field of Thinking Machines: Inkling Small (batch) is on its tool page: open its card above.
What can you do with Inkling Small (batch) in Muse Code?
Turn reproducible failures and written requirements into code changes with Inkling Small. Ask the agent to inspect the relevant files, make a focused patch, and verify the result with tests.
Repair a reproducible bug
Ask Inkling Small to trace a failing test, propose a focused code change, and check the result against the original failure.
Grade answers against a rubric
Give Inkling Small the reference answer and scoring rules. Request a score, supporting evidence, and a reason for each deduction.
Draft synthetic training examples
Use Inkling Small to draft labeled examples from a written policy, including ambiguous cases for human review.
Answer questions from retrieved passages
Provide Inkling Small with relevant documentation excerpts. Require passage citations and an explicit statement when the evidence does not answer the question.
Compare a specification with an implementation
Have Inkling Small map written requirements to code behavior and identify missing cases, with references to the supplied evidence.
Draft localized support replies
Give Inkling Small approved support guidance and a terminology glossary. Ask for a reply in the customer's language, then review meaning and terminology.
More on Inkling Small (batch) and Muse Code
Overview
Inkling Small (batch) in other agents
More Muse Code guides
Questions about Inkling Small (batch) in Muse Code
Can Muse Code use Inkling Small (batch)?
Yes. Add Spiral to Muse Code as an MCP server, and Muse Code can find and run Inkling Small (batch) like any built-in tool.
How much does Inkling Small (batch) cost in Muse Code?
Up to $0.005129 per run at the default inputs. $0.468 per 1M input tokens, $1.248 per 1M output tokens. You pay from your Spiral balance as you go, with no subscription, and a failed run is not charged. Your Muse Code plan is separate.
Do I need an OpenRouter account to use Inkling Small (batch) in Muse Code?
No. Spiral runs Inkling Small (batch) on its own OpenRouter account. You sign in to Spiral once and never paste a key.
How long does it take to set up Inkling Small (batch) in Muse Code?
About two minutes: install Muse Code, then ask Muse to add the server, then sign in.
Does Muse Code wait while Inkling Small (batch) runs?
Inkling Small (batch) runs in the background. Muse Code starts the run, checks on it and gives you the text, json when it's ready.
What is Inkling Small good at?
Coding, reasoning, grading against a rubric, and generating synthetic training examples are supported uses. Give it a concrete task and an explicit standard for a correct answer.
How does Inkling Small compare with Inkling?
The maker reports stronger reasoning and agentic coding benchmark results for Inkling Small. Inkling retains an advantage in knowledge coverage and factuality.
How much context can Inkling Small handle?
Thinking Machines specifies a context window of up to one million tokens for the model. For document analysis, ask for supporting passages and identify which documents should take priority.
Does more thinking always improve the answer?
No. Higher effort generally encourages more reasoning, but the maker says it does not guarantee greater accuracy on every response. Compare results on representative tasks before choosing an effort level.
Can Inkling Small work in languages besides English?
Yes, the maker documents general multilingual support. Test the language and terminology you need with real examples before using the results.
Start using Inkling Small (batch) in Muse Code
One Spiral account for every tool your agent needs. No subscription; you pay per run.
Paste this into Muse Code:
set up https://spiral.ai/SKILL.md


