How to Use Kimi K2.5 in Hermes
Kimi K2.5 suits visual coding, software maintenance, and work across long documents. It combines image understanding with Instant and Thinking modes.
Prompts to try with Kimi K2.5 in Hermes
Copy an example into Hermes, or ask in your own words.
Every weekday, use Kimi K2.5 to summarize the collected release announcements. Separate new behavior, breaking changes, and unresolved questions, and include a source link for each finding.
Each week, use Kimi K2.5 to compare the collected product announcements with last week's summary. Report substantive changes and cite the passages that establish them.
Every morning, use Kimi K2.5 to organize the collected support reports into recurring problems. Include representative evidence and distinguish reported symptoms from inferred causes.
Each Friday, use Kimi K2.5 to turn this week's collected project updates into a one-page status brief. List completed work, conflicting reports, and questions requiring follow-up.
Each month, use Kimi K2.5 with Thinking enabled to compare the collected proposals against our stated evaluation criteria. Show the evidence for each assessment and flag missing information.
How to set up Kimi K2.5 in Hermes
Pick the way that suits you. Each takes about two minutes.
Use the Spiral skill
- Send Hermes 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
- Add the server and sign in
Hermes opens your browser: sign in to Spiral and click Authorize. No API key. When Hermes asks Enable all tools? \[Y/n/select], answer Y: n cancels without saving the server, and select hides the tools you leave out.
hermes mcp add spiral \ --url https://api.spiral.ai/mcp \ --auth oauth \ --connect-timeout 300
- Check it
In a chat that was already open, run /reload-mcp.
hermes mcp list hermes mcp test spiral
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 K2.5
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-k2.5", "input": { "messages": [ { "role": "system" } ] } }'
Then paste one of the prompts above: Hermes finds Kimi K2.5 on Spiral, shows the price and runs it.
Kimi K2.5 providers and prices
Spiral runs Kimi K2.5 through OpenRouter. Open a card for today's price, with every fee included.
Up to $0.013336 per run at the default inputs. $0.65 per 1M input tokens, $3.25 per 1M output tokens.
Settings you can ask for
Ask Hermes 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 | ||
| Logprobs | on or off | ||
| Max tokens | 1 to 235929 | The most output tokens to generate. | |
| Temperature | 0 to 2 | ||
| Top logprobs | 0 to 20 | ||
| Presence penalty | -2 to 2 | ||
| Frequency penalty | -2 to 2 | ||
| Repetition penalty | 0 to 2 |
Every API field of MoonshotAI: Kimi K2.5 is on its tool page: open its card above.
What can you do with Kimi K2.5 in Hermes?
Turn recurring batches of reports and research into consistent briefs with Kimi K2.5. Define the comparison period, evidence requirements, and conditions that deserve attention.
Implement a design reference
Use Kimi K2.5 to turn a screenshot into interface code, with explicit requirements for layout, responsive behavior, and working controls.
Investigate visual bugs
Give Kimi K2.5 the intended design and a screenshot of the current result. Request a prioritized list of mismatches and proposed fixes.
Refactor existing code
Ask Kimi K2.5 to simplify a module while preserving its documented behavior, then explain the changes and remaining risks.
Draft regression tests
Have Kimi K2.5 turn a bug report and relevant code into a focused regression test and a proposed repair.
Compare long documents
Ask Kimi K2.5 to compare labeled source texts, identify disagreements, and tie each finding to a supporting passage.
Prepare an evidence-based brief
Use Kimi K2.5 to organize collected research into findings, evidence, open questions, and a recommendation for a named audience.
Plan spreadsheet analysis
Give Kimi K2.5 a business question and a description of the available data. Request proposed calculations, table layouts, and checks for missing data.
Draft a presentation
Ask Kimi K2.5 to turn supplied research into a slide outline with one claim per slide, supporting evidence, and speaker notes.
More on Kimi K2.5 and Hermes
Overview
Kimi K2.5 in other agents
More Hermes guides
Questions about Kimi K2.5 in Hermes
Can Hermes use Kimi K2.5?
Yes. Add Spiral to Hermes as an MCP server, and Hermes can find and run Kimi K2.5 like any built-in tool.
How much does Kimi K2.5 cost in Hermes?
Up to $0.013336 per run at the default inputs. $0.65 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 Hermes plan is separate.
Do I need an OpenRouter account to use Kimi K2.5 in Hermes?
No. Spiral runs Kimi K2.5 on its own OpenRouter account. You sign in to Spiral once and never paste a key.
How long does it take to set up Kimi K2.5 in Hermes?
About two minutes: add the server and sign in, then check it.
Does Hermes wait while Kimi K2.5 runs?
Kimi K2.5 runs in the background. Hermes starts the run, checks on it and gives you the text, json when it's ready.
What is Kimi K2.5 good at?
Moonshot emphasizes visual coding, software maintenance, and work across dense documents. Practical uses include implementing a screenshot, investigating a layout bug, and turning source material into a report.
How much context can Kimi K2.5 handle?
The documented context window is 256K tokens. For long tasks, ask for a source map first and work through named sections so you can check what the answer covers.
What are Instant and Thinking modes?
Kimi K2.5 supports both modes. Try Thinking for a difficult debugging or analysis task, and compare Instant on routine extraction or rewriting using the same examples.
Which temperature should I use?
Moonshot recommends 1.0 for Thinking and 0.6 for Instant. Start with those values and judge changes against representative tasks.
Where does Kimi K2.5 need extra checking?
Factual recall deserves scrutiny: Moonshot reports 36.9% on SimpleQA Verified. Ask for evidence, check citations, and require an explicit uncertainty statement when the supplied material does not support an answer.
Start using Kimi K2.5 in Hermes
One Spiral account for every tool your agent needs. No subscription; you pay per run.
Paste this into Hermes:
set up https://spiral.ai/SKILL.md


