[Integrations](https://spiral.ai/integrations) › [Claude Code](https://spiral.ai/integrations/claude-code) › [Gemini Flash Latest](https://spiral.ai/integrations/claude-code/gemini-flash-latest)

# How to Use Gemini Flash Latest in Claude Code

Use Gemini Flash Latest for coding, document analysis, and structured extraction. Unlike a fixed Gemini 3.8 Flash selection, it follows new Flash releases, so behavior and capabilities can change.

|  |  |
| --- | --- |
| Model | Gemini Flash Latest |
| Available through | Vertex AI |
| Category | LLM |
| Price | Up to $0.040005 per run at the default inputs. $1.95 per 1M input tokens, $0.0975 per 1M cached input tokens, $0.108333335 per 1M cache write tokens, $9.75 per 1M output tokens. |
| Returns | text, json |
| Runs | In the background |
| Setup | About 2 minutes, no API key |

## Prompts to try with Gemini Flash Latest in Claude Code

- Use Gemini Flash Latest with high reasoning effort to investigate the failing test in this project. Trace the relevant code, make the smallest justified fix, and run the affected tests.
- Use Gemini Flash Latest to review the current diff for behavior changes and missed edge cases. Report each finding with its file, supporting evidence, and a suggested regression test.
- Use Gemini Flash Latest to compare the project's implementation with its documented requirements. Return a table of requirement, implementation location, and any gap you can demonstrate.
- Use Gemini Flash Latest to turn the project's sample support messages into a proposed classification dataset. Follow existing labels, preserve the original text, and flag ambiguous examples.
- Use Gemini Flash Latest to review this project's JSON extraction logic. Check how it handles missing dates, conflicting names, and valid JSON containing incorrect values, then propose focused improvements.

## How to set up Gemini Flash Latest in Claude Code

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

### Use Spiral's MCP server

1. **Add the hosted server**: This adds Spiral to the current project. Add `--scope user` to use it in every project.

```bash
claude mcp add --transport http spiral https://api.spiral.ai/mcp
```

2. **Sign in**: Start `claude`, type `/mcp`, choose spiral and Authenticate. Sign in to Spiral in the browser and click Authorize. No API key.

### Use the Spiral skill

1. **Send Claude Code 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 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 Gemini Flash Latest**: 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": "vertex",
    "model": "~google/gemini-flash-latest",
    "input": {
      "messages": [
        {
          "role": "system"
        }
      ]
    }
  }'
```

Then paste one of the prompts above: Claude Code finds Gemini Flash Latest on Spiral, shows the price and runs it.

Setup guide in the docs: https://spiral.ai/docs/quickstarts/claude-code

## Gemini Flash Latest tools and prices

- [Gemini Flash Latest](https://spiral.ai/tools/vertex/~google/gemini-flash-latest) (Vertex AI): Up to $0.040005 per run at the default inputs. $1.95 per 1M input tokens, $0.0975 per 1M cached input tokens, $0.108333335 per 1M cache write tokens, $9.75 per 1M output tokens.

## Settings you can ask for

| Setting | Allowed values | Default | Description |
| --- | --- | --- | --- |
| seed |  | — |  |
| stop |  | — |  |
| tools |  | — |  |
| reasoning |  | — | Reasoning effort; reasoning tokens count toward max_tokens. |
| max_tokens | 1 to 65536 | — | 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). |
| tool_choice |  | — |  |

## What can you do with Gemini Flash Latest in Claude Code?

Turn bug reports and project code into proposed fixes, reviews, and test plans with Gemini Flash Latest. Give it the relevant files, expected behavior, and a concrete definition of success.

### Bug investigation

Use Gemini Flash Latest to compare an error report with relevant code, identify likely causes, and propose a focused fix with a regression check.

### Code review

Ask Gemini Flash Latest to inspect a proposed change for incorrect assumptions, missed edge cases, and behavior that needs testing.

### Document comparison

Give Gemini Flash Latest the text of two policy or specification versions and request a table of changed requirements with supporting excerpts.

### Support request classification

Use Gemini Flash Latest to assign incoming messages to defined categories, following labeled examples and flagging ambiguous requests for review.

### Structured record extraction

Ask Gemini Flash Latest to extract names, dates, and requested actions from message text. Check each returned value against its source.

### Meeting summaries

Turn meeting transcript text into decisions, open questions, and action items with Gemini Flash Latest. Require evidence for owners and deadlines.

### Multilingual response drafts

Use Gemini Flash Latest to draft responses in a customer's language from supplied facts, preserving product names and approved terminology.

## Overview

- [Google Vertex AI](https://spiral.ai/providers/vertex)
- [All Claude Code guides](https://spiral.ai/integrations/claude-code)

## Gemini Flash Latest in other agents

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

## More Claude Code guides

- [Gemini 4 Neon](https://spiral.ai/integrations/claude-code/gemini-4-neon)
- [Gemini Pro Latest](https://spiral.ai/integrations/claude-code/gemini-pro-latest)
- [Gemini 4 Hydrogen Vertex AI](https://spiral.ai/integrations/claude-code/gemini-4-hydrogen-vertex-ai)
- [Gemma 2 27B Instruct](https://spiral.ai/integrations/claude-code/gemma-2-27b-instruct)
- [Gemini 4 Hydrogen](https://spiral.ai/integrations/claude-code/gemini-4-hydrogen)
- [Gemma 3 12B](https://spiral.ai/integrations/claude-code/gemma-3-12b)

## Questions about Gemini Flash Latest in Claude Code

**Can Claude Code use Gemini Flash Latest?**

Yes. Add Spiral to Claude Code as an MCP server, and Claude Code can find and run Gemini Flash Latest like any built-in tool.

**How much does Gemini Flash Latest cost in Claude Code?**

Up to $0.040005 per run at the default inputs. $1.95 per 1M input tokens, $0.0975 per 1M cached input tokens, $0.108333335 per 1M cache write tokens, $9.75 per 1M output tokens. You pay from your Spiral balance as you go, with no subscription, and a failed run is not charged. Your Claude Code plan is separate.

**Do I need a Vertex AI account to use Gemini Flash Latest in Claude Code?**

No. Spiral runs Gemini Flash Latest on its own Vertex AI account. You sign in to Spiral once and never paste a key.

**How long does it take to set up Gemini Flash Latest in Claude Code?**

About two minutes: add the hosted server, then sign in.

**Does Claude Code wait while Gemini Flash Latest runs?**

Gemini Flash Latest runs in the background. Claude Code starts the run, checks on it and gives you the text, json when it's ready.

**Does Gemini Flash Latest always use the same release?**

No. Google updates the alias as new Flash releases arrive, so behavior and capabilities can change. Recheck a small set of representative tasks after an update.

**What is Gemini Flash useful for?**

Use it for code changes, document analysis, and tasks that turn unstructured text into organized records. Google's Gemini 3.8 Flash documentation specifically emphasizes software engineering, autonomous agents, and complex enterprise work.

**How much text can Gemini Flash handle?**

For Gemini 3.8 Flash, Google lists 1,048,576 input tokens and up to 65,536 output tokens. Those limits describe that release; check the release behind Gemini Flash Latest before relying on them.

**Which languages does Gemini Flash support?**

Google's general-use Gemini models understand and respond in over 100 languages, including English, Spanish, French, German, Japanese, and Arabic. For translation work, specify the target language, audience, and terms to preserve.

**How do I get more consistent answers from Gemini Flash?**

Provide examples that show the format and level of detail you want. For extraction or classification, include examples of missing information and ambiguous cases.

