[Integrations](https://spiral.ai/integrations) › [Hermes](https://spiral.ai/integrations/hermes) › [MiMo-V2.6 Flash](https://spiral.ai/integrations/hermes/mimo-v2-6-flash)

# How to Use MiMo-V2.6 Flash in Hermes

MiMo-V2.6 Flash suits repeated office tasks, code work, and long-document analysis. Xiaomi positions Flash for frequent professional tasks and MiMo-V2.6 Pro for complex projects and longer assignments.

|  |  |
| --- | --- |
| Model | MiMo-V2.6 Flash |
| Available through | OpenRouter |
| Category | LLM |
| Price | Up to $0.001498 per run at the default inputs. $0.182 per 1M input tokens, $0.364 per 1M output tokens. |
| Returns | text, json |
| Runs | In the background |
| Setup | About 2 minutes, no API key |

## Prompts to try with MiMo-V2.6 Flash in Hermes

- Each morning, use MiMo-V2.6 Flash to summarize the project updates supplied since the previous run. Report new decisions, blockers, and next actions with source references. Omit unchanged items and keep the brief under 400 words.
- Each day, use MiMo-V2.6 Flash to classify newly supplied support messages by issue and reported impact. Return one JSON record per message, preserve its reference, and mark missing details as null.
- Each week, use MiMo-V2.6 Flash to compare the current project status text with the previous summary. List resolved blockers, new risks, and changed commitments. Include supporting passages and do not infer completion from silence.
- After each scheduled test run, use MiMo-V2.6 Flash to summarize the supplied failures. Group failures that appear to share a cause, distinguish evidence from hypotheses, and suggest the next diagnostic step. Do not rerun identical checks without new information.
- Every Friday, use MiMo-V2.6 Flash to turn the week's supplied progress notes into a six-slide review outline. Cover completed work, evidence, blockers, decisions needed, and next week's priorities. Leave unsupported claims out.

## How to set up MiMo-V2.6 Flash in Hermes

Pick the way that suits you. Each takes about two minutes.

### Use the Spiral skill

1. **Send Hermes this message**: Your agent reads Spiral's skill and sets itself up.

```
set up https://spiral.ai/SKILL.md
```

2. **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

1. **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.

```bash
hermes mcp add spiral \
  --url https://api.spiral.ai/mcp \
  --auth oauth \
  --connect-timeout 300
```

2. **Check it**: In a chat that was already open, run `/reload-mcp`.

```bash
hermes mcp list
hermes mcp test spiral
```

### Use the Spiral CLI

1. **Install and sign in**: Installs the Spiral CLI, signs you in and adds its skill to your coding agents.

```bash
npx -y spiral-cli@latest init
```

### Call the Spiral API

1. **Create an API key**: In the console, under API keys.

2. **Run MiMo-V2.6 Flash**: One request; the result comes back by polling or a webhook.

```bash
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": "xiaomi/mimo-v2.6-flash",
    "input": {
      "messages": [
        {
          "role": "system"
        }
      ]
    }
  }'
```

Then paste one of the prompts above: Hermes finds MiMo-V2.6 Flash on Spiral, shows the price and runs it.

Setup guide in the docs: https://spiral.ai/docs/quickstarts/hermes

## MiMo-V2.6 Flash tools and prices

- [Xiaomi: MiMo-V2.6-Flash](https://spiral.ai/tools/openrouter/xiaomi/mimo-v2.6-flash) (OpenRouter): Up to $0.001498 per run at the default inputs. $0.182 per 1M input tokens, $0.364 per 1M output tokens.

## Settings you can ask for

| Setting | Allowed values | Default | Description |
| --- | --- | --- | --- |
| seed |  | — |  |
| stop |  | — |  |
| min_p | 0 to 1 | — |  |
| tools |  | — |  |
| top_k | 0 to … | — |  |
| top_p | 0 to 1 | — |  |

## What can you do with MiMo-V2.6 Flash in Hermes?

Turn incoming updates into recurring briefs and structured queues with MiMo-V2.6 Flash. Define what changed, what evidence to include, and when the agent should stop.

### Fix a reproducible software bug

Give MiMo-V2.6 Flash the failing behavior, relevant code, and expected result. Ask for a focused patch, a regression test, and a report of what passed.

### Add tests around an existing function

Ask MiMo-V2.6 Flash to identify boundary cases in supplied code and write tests using the project's existing conventions.

### Build an interface from a written brief

Describe the required screens, content, and interactions. Ask MiMo-V2.6 Flash to produce the front-end code and check each requested behavior.

### Compare lengthy project documents

Supply labeled specifications or reports and ask MiMo-V2.6 Flash to extract agreements, contradictions, and unanswered questions with supporting passages.

### Extract structured records from text

Turn supplied support messages or operational notes into JSON records. Define the required keys and ask MiMo-V2.6 Flash to use null for missing information.

### Prepare a recurring operations brief

Have MiMo-V2.6 Flash summarize newly supplied updates into decisions, blockers, owners, and next actions. Require evidence for each item and omit unchanged updates.

### Draft a presentation from source notes

Ask MiMo-V2.6 Flash to turn a written brief into a slide outline with one claim per slide, supporting evidence, and speaker notes.

## Overview

- [OpenRouter](https://spiral.ai/providers/openrouter)
- [All Hermes guides](https://spiral.ai/integrations/hermes)

## MiMo-V2.6 Flash in other agents

- [Amp](https://spiral.ai/integrations/amp/mimo-v2-6-flash)
- [ChatGPT](https://spiral.ai/integrations/chatgpt/mimo-v2-6-flash)
- [Claude](https://spiral.ai/integrations/claude-desktop/mimo-v2-6-flash)
- [Claude Code](https://spiral.ai/integrations/claude-code/mimo-v2-6-flash)
- [Cline](https://spiral.ai/integrations/cline/mimo-v2-6-flash)
- [Codex](https://spiral.ai/integrations/codex/mimo-v2-6-flash)
- [Cursor](https://spiral.ai/integrations/cursor/mimo-v2-6-flash)
- [Devin](https://spiral.ai/integrations/devin/mimo-v2-6-flash)
- [dots](https://spiral.ai/integrations/dots/mimo-v2-6-flash)
- [Factory Droid](https://spiral.ai/integrations/factory-droid/mimo-v2-6-flash)
- [Gemini](https://spiral.ai/integrations/gemini/mimo-v2-6-flash)
- [Gemini CLI](https://spiral.ai/integrations/gemini-cli/mimo-v2-6-flash)
- [Google Antigravity](https://spiral.ai/integrations/antigravity/mimo-v2-6-flash)
- [goose](https://spiral.ai/integrations/goose/mimo-v2-6-flash)
- [Grok](https://spiral.ai/integrations/grok/mimo-v2-6-flash)
- [Grok Build](https://spiral.ai/integrations/grok-build/mimo-v2-6-flash)
- [Instinct](https://spiral.ai/integrations/instinct/mimo-v2-6-flash)
- [Manus](https://spiral.ai/integrations/manus/mimo-v2-6-flash)
- [Muse](https://spiral.ai/integrations/muse/mimo-v2-6-flash)
- [Muse Code](https://spiral.ai/integrations/muse-code/mimo-v2-6-flash)
- [n8n](https://spiral.ai/integrations/n8n/mimo-v2-6-flash)
- [Notion](https://spiral.ai/integrations/notion/mimo-v2-6-flash)
- [OpenClaw](https://spiral.ai/integrations/openclaw/mimo-v2-6-flash)
- [opencode](https://spiral.ai/integrations/opencode/mimo-v2-6-flash)
- [Perplexity](https://spiral.ai/integrations/perplexity/mimo-v2-6-flash)
- [Raycast](https://spiral.ai/integrations/raycast/mimo-v2-6-flash)
- [VS Code](https://spiral.ai/integrations/vscode/mimo-v2-6-flash)
- [Windsurf](https://spiral.ai/integrations/windsurf/mimo-v2-6-flash)
- [Zapier](https://spiral.ai/integrations/zapier/mimo-v2-6-flash)

## More Hermes guides

- [MiMo-V2.5-Pro](https://spiral.ai/integrations/hermes/mimo-v2-5-pro)
- [MiniMax M2.5](https://spiral.ai/integrations/hermes/minimax-m2-5)
- [MiMo-V2.5](https://spiral.ai/integrations/hermes/mimo-v2-5)
- [MiniMax M2.7](https://spiral.ai/integrations/hermes/minimax-m2-7)
- [MiMo V2.6 Pro UltraSpeed](https://spiral.ai/integrations/hermes/mimo-v2-6-pro-ultraspeed)
- [MiniMax M3](https://spiral.ai/integrations/hermes/minimax-m3)

## Questions about MiMo-V2.6 Flash in Hermes

**Can Hermes use MiMo-V2.6 Flash?**

Yes. Add Spiral to Hermes as an MCP server, and Hermes can find and run MiMo-V2.6 Flash like any built-in tool.

**How much does MiMo-V2.6 Flash cost in Hermes?**

Up to $0.001498 per run at the default inputs. $0.182 per 1M input tokens, $0.364 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 MiMo-V2.6 Flash in Hermes?**

No. Spiral runs MiMo-V2.6 Flash on its own OpenRouter account. You sign in to Spiral once and never paste a key.

**How long does it take to set up MiMo-V2.6 Flash in Hermes?**

About two minutes: add the server and sign in, then check it.

**Does Hermes wait while MiMo-V2.6 Flash runs?**

MiMo-V2.6 Flash runs in the background. Hermes starts the run, checks on it and gives you the text, json when it's ready.

**What is MiMo-V2.6 Flash good for?**

Xiaomi positions it for frequent professional tasks. Useful starting points include code changes, document summaries, structured extraction, and drafting interfaces from written requirements.

**How much context can MiMo-V2.6 Flash handle?**

Xiaomi documents a 1 million-token context window. For document analysis, label each source and ask for supporting passages so you can check the result.

**How long can a MiMo-V2.6 Flash answer be?**

The documented maximum output is 128K tokens. Reasoning shares the completion budget with the visible answer, so 128K is not a promise of 128K tokens of finished prose.

**Does lowering temperature change its thinking-mode answers?**

Xiaomi says thinking mode overrides custom sampling settings with a temperature of 1.0 and a nucleus-sampling threshold of 0.95. Specify the evidence, format, and length you want in the prompt.

**What weakness should I watch for in tool-heavy tasks?**

Xiaomi documented repeated tool calls that consumed context without advancing the task, then released an update that substantially reduced the behavior. Give each task a clear stopping condition and ask the agent to report a blocker when repeated actions produce no new information.

