How to Use Kimi K2 0905 in Instinct
Use Kimi K2 0905 for longer coding tasks and frontend work. It expands Kimi K2 0711's 128K context window to 256K and improves frontend coding.
Prompts to try with Kimi K2 0905 in Instinct
Copy an example into Instinct, or ask in your own words.
Use Kimi K2 0905 on each scheduled run to review newly retrieved release notes. Summarize breaking changes, cite the relevant passages, and list questions the notes leave unanswered.
Use Kimi K2 0905 each week to investigate one unresolved engineering question from the supplied queue. Use the available search tools and return a brief with sources and remaining uncertainties.
Use Kimi K2 0905 each Friday to draft English and Chinese engineering updates from the supplied completion records. Preserve issue identifiers and exclude work without supporting evidence.
Every Monday at 9am: ask Kimi K2 0905 through Spiral to review the diff on this branch and list the three riskiest changes.
Get a second opinion from Kimi K2 0905 on this function, then tell me what the call cost.
How to set up Kimi K2 0905 in Instinct
Pick the way that suits you. Each takes about two minutes.
Use the Spiral skill
- Send Instinct 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 an API key
In the console, open API keys and create a key for Instinct. Copy it.
- Send Instinct this message
Instinct reads Spiral's skill and asks for your API key.
set up https://spiral.ai/SKILL.md
- Give Instinct the key
When Instinct asks for the key, add it where Instinct keeps secrets, never in the chat itself.
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 Kimi K2 0905
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": "moonshotai/kimi-k2-0905", "input": { "messages": [ { "role": "system" } ] } }'
Then paste one of the prompts above: Instinct finds Kimi K2 0905 on Spiral, shows the price and runs it.
Kimi K2 0905 providers and prices
Spiral runs Kimi K2 0905 through OpenRouter. Open a card for today's price, with every fee included.
Up to $0.01334 per run at the default inputs. $0.78 per 1M input tokens, $3.25 per 1M output tokens.
Settings you can ask for
Ask Instinct for any of these in your prompt.
| Setting | Allowed values | Default | Description |
|---|---|---|---|
| Seed | a whole number | ||
| Top k | a whole number from 0 | ||
| Top p | 0 to 1 | ||
| Max tokens | 1 to 98304 | The most output tokens to generate. | |
| Temperature | 0 to 2 | ||
| Presence penalty | -2 to 2 | ||
| Frequency penalty | -2 to 2 | ||
| Repetition penalty | 0 to 2 |
Every API field of MoonshotAI: Kimi K2 0905 is on its tool page: open its card above.
What can you do with Kimi K2 0905 in Instinct?
Turn recurring text inputs into technical briefs and support drafts with Kimi K2. Give each run a fixed question and require evidence for its findings.
Fix a reproducible bug
Provide the failing behavior and expected result. Ask for a small patch, run the relevant checks, and use the failures to guide revisions.
Build a frontend feature
Describe the interface in text, including empty states and interactions. Implement one component at a time; Kimi K2 0905 has documented frontend improvements.
Refactor existing code
Specify the behavior that must stay the same, make a bounded change, and compare test results before continuing.
Develop a data analysis script
Ask for code that calculates a defined metric, inspect the execution results, and request an explanation tied to those results.
Investigate a technical question
Set one question, gather relevant text through available tools, and ask for a brief that separates evidence from unresolved points.
Draft English or Chinese responses
Supply the facts, audience, and desired language. Request a short draft with technical terms and names preserved.
Prepare support replies
Have the agent retrieve the relevant policy or product information, then draft a reply based on the returned evidence.
Turn a specification into tested code
Break the specification into observable behaviors. Implement and check each behavior before combining the pieces.
More on Kimi K2 0905 and Instinct
Overview
Kimi K2 0905 in other agents
More Instinct guides
Questions about Kimi K2 0905 in Instinct
Can Instinct use Kimi K2 0905?
Yes. Add Spiral to Instinct as an MCP server, and Instinct can find and run Kimi K2 0905 like any built-in tool.
How much does Kimi K2 0905 cost in Instinct?
Up to $0.01334 per run at the default inputs. $0.78 per 1M input tokens, $3.25 per 1M output tokens. You pay from your Spiral balance as you go, with no subscription, and a failed run is not charged. Your Instinct plan is separate.
Do I need an OpenRouter account to use Kimi K2 0905 in Instinct?
You need no OpenRouter account. Instinct uses one Spiral API key for every tool, and Spiral runs Kimi K2 0905 on its own OpenRouter account.
How long does it take to set up Kimi K2 0905 in Instinct?
About two minutes: create an API key, then send Instinct this message, then give Instinct the key.
Does Instinct wait while Kimi K2 0905 runs?
Kimi K2 0905 runs in the background. Instinct starts the run, checks on it and gives you the text, json when it's ready.
What is Kimi K2 good for?
Coding, tool-assisted problem solving, and text questions are useful starting points. For software work, ask for a change, test it, and return the results for another pass.
Can Kimi K2 understand screenshots or images?
Kimi K2 does not support native vision. Provide extracted text or a written description when asking about visual material.
Does Kimi K2 use extended thinking?
Kimi K2 0711 and Kimi K2 0905 are non-thinking models. They can solve reasoning tasks, but they do not provide the extended thinking behavior of Kimi K2 Thinking.
What temperature should I start with?
Moonshot recommends 0.6 for Kimi K2 0711 and Kimi K2 0905. Use that as a starting point, then compare results on your actual task.
How should I prompt Kimi K2 for coding?
Give it a specific behavior to change and a way to check success. For larger projects, work through small changes with test feedback instead of requesting the entire application in one answer.
Start using Kimi K2 0905 in Instinct
One Spiral account for every tool your agent needs. No subscription; you pay per run.
Paste this into Instinct:
set up https://spiral.ai/SKILL.md


