How to Use Ember-1 in n8n
Ember-1 is built for coding, document analysis, and tasks that require several reasoning steps. Built on Kimi K3, it uses about 40% fewer reasoning tokens while maintaining comparable quality in Fireworks' evaluations.
Prompts to try with Ember-1 in n8n
Copy an example into n8n, or ask in your own words.
Every morning, use Ember-1 to compare the supplied product updates with yesterday's summary. Report changed behavior, unresolved questions, and supporting passages in under 400 words.
Each week, use Ember-1 to summarize the supplied customer feedback. Group recurring problems, include representative excerpts, and distinguish reported behavior from proposed explanations.
Every evening, use Ember-1 to review the supplied incident summaries. Draft a list of recurring symptoms and questions for investigation, keeping observations separate from suspected causes.
Each Monday, use Ember-1 to compare the supplied project status updates with the previous week's updates. List changed deadlines, new blockers, and conflicting statements with source labels.
Each Friday, use Ember-1 to check the calculations in the supplied weekly metrics summary. Show discrepancies with equations and units, and flag missing inputs.
How to set up Ember-1 in n8n
Pick the way that suits you. Each takes about two minutes.
Use the Spiral skill
- Send n8n 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 Spiral's MCP server
- Create a key
Create a key under API keys in the console.
- Add an MCP Client node
In a workflow, add an MCP Client node (or, under an AI Agent node, an MCP Client Tool) and set:
https://api.spiral.ai/mcp
- Pick a tool
The node lists Spiral's tools. In an MCP Client node pick one, such as run, and give it its arguments:
{ "provider": "fal", "model": "fal-ai/flux/schnell", "input": { "prompt": "a lighthouse at dawn" }, "wait": true } - Run Ember-1 in a workflow step
Give the step an instruction like the ones below and run it on your trigger or schedule. Spiral finds Ember-1, prices it and runs it each time the step fires.
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 Ember-1
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": "fireworks/ember-1", "input": { "messages": [ { "role": "system" } ] } }'
Then paste one of the prompts above: n8n finds Ember-1 on Spiral, shows the price and runs it.
Ember-1 providers and prices
Spiral runs Ember-1 through OpenRouter. Open a card for today's price, with every fee included.
Up to $0.080009 per run at the default inputs. $3.90 per 1M input tokens, $19.50 per 1M output tokens.
Settings you can ask for
Ask n8n for any of these in your prompt.
| Setting | Allowed values | Default | Description |
|---|---|---|---|
| Top k | a whole number from 0 | ||
| Top p | 0 to 1 | ||
| Logprobs | on or off | ||
| Max tokens | 1 to 943718 | The most output tokens to generate. | |
| Temperature | 0 to 2 | ||
| Top logprobs | 0 to 20 | ||
| Presence penalty | -2 to 2 | ||
| Reasoning effort | none, low, high, or max | max | How much the model reasons. |
| Frequency penalty | -2 to 2 | ||
| Repetition penalty | 0 to 2 |
Every API field of Fireworks: Ember-1 is on its tool page: open its card above.
What can you do with Ember-1 in n8n?
Use Ember-1 to turn recurring batches of supplied information into comparisons and review drafts. Define what counts as a meaningful change and require supporting evidence in each report.
Fix a reproducible bug
Give Ember-1 the failing behavior, relevant code, and test output. Request a focused patch and a regression test tied to the reported failure.
Review a proposed code change
Ask Ember-1 to compare a patch with its requirements and identify correctness risks. Require a concrete failure scenario for each finding.
Plan a refactor
Supply the current implementation and constraints. Ask Ember-1 for a staged refactor plan with behavior checks after each step.
Compare long documents
Provide labeled documents and ask Ember-1 to extract agreements, contradictions, and unanswered questions, with section references for each finding.
Prepare a research brief
Give Ember-1 collected source material and a decision to support. Request a brief that separates sourced findings, assumptions, and missing evidence.
Explain an image
Supply an image and ask Ember-1 to describe its visible subjects and setting. Request that uncertain details be identified explicitly.
Check a calculation
Provide the quantities, units, and rules. Ask Ember-1 to calculate the result, show the key equations, and flag missing inputs.
More on Ember-1 and n8n
Overview
Ember-1 in other agents
Questions about Ember-1 in n8n
Can n8n use Ember-1?
Yes. Add Spiral to n8n as an MCP server, and n8n can find and run Ember-1 like any built-in tool.
How much does Ember-1 cost in n8n?
Up to $0.080009 per run at the default inputs. $3.90 per 1M input tokens, $19.50 per 1M output tokens. You pay from your Spiral balance as you go, with no subscription, and a failed run is not charged. Your n8n plan is separate.
Do I need an OpenRouter account to use Ember-1 in n8n?
You need no OpenRouter account. n8n uses one Spiral API key for every tool, and Spiral runs Ember-1 on its own OpenRouter account.
How long does it take to set up Ember-1 in n8n?
About two minutes: create a key, then add an MCP Client node, then pick a tool.
Does n8n wait while Ember-1 runs?
Ember-1 runs in the background. n8n starts the run, checks on it and gives you the text, json when it's ready.
What is Ember-1 best suited for?
Coding, knowledge work, and tasks that require several reasoning steps are its stated uses. Practical starting points include debugging code, comparing supplied documents, and turning research into a concise brief.
How does Ember-1 differ from Kimi K3?
Ember-1 builds on Kimi K3 and was trained to shorten its reasoning traces. Fireworks reports approximately 40% fewer tokens with comparable quality across its evaluations.
How much context can Ember-1 handle?
Fireworks lists approximately 1.04 million tokens of context. For document comparisons, label each source and request references to the sections behind each conclusion.
Can Ember-1 analyze images?
Yes. Ember-1 accepts images alongside text and returns text, so you can request a description or ask questions about visible content.
Does shorter reasoning mean Ember-1 always matches Kimi K3?
No. Fireworks reported 92.2% on SWE-bench Verified for Ember-1 versus 93.2% for Kimi K3 at maximum reasoning effort. Treat comparable quality as a benchmark finding and check generated code against your own tests.
Start using Ember-1 in n8n
One Spiral account for every tool your agent needs. No subscription; you pay per run.
Paste this into n8n:
set up https://spiral.ai/SKILL.md


