Spiral gives your agent every tool with one account. Browse the tools →

Claude Code · LLMReleased 2026-03-31

How to Use Grok 4.20 in Claude Code

Grok 4.20 fits document analysis, coding, and questions that require several logical steps. Enable reasoning for difficult problems; disable it for straightforward extraction or routine Q&A.

Prompts to try with Grok 4.20 in Claude Code

Copy an example into Claude Code, or ask in your own words.

Use Grok 4.20 with reasoning enabled to investigate the failing test in this project. Trace the relevant code, identify the most likely cause, make the smallest justified fix, and run the relevant tests.

Grok 4.20Claude Code
Try asking
  • Use Grok 4.20 to review this branch's diff against the requirements in the project docs. Report concrete defects with file references, a triggering example, and a proposed correction.

  • Use Grok 4.20 with reasoning enabled to review the database migration in this project. Check ordering, compatibility, rollback behavior, and missing acceptance checks before proposing changes.

  • Use Grok 4.20 to create a JSON extraction schema for the invoice text examples in this repository. Allow missing values explicitly, preserve source text where needed, and validate the extracted records against the schema.

  • Use Grok 4.20 to draft Spanish translations for this project's English support messages. Follow the existing glossary, preserve interpolation tokens, and flag phrases that need human review.

How to set up Grok 4.20 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.

    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.

    npx -y spiral-cli@latest init

Call the Spiral API

  1. Create an API key

    In the console, under API keys.

  2. Run Grok 4.20

    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": "x-ai/grok-4.20",
        "input": {
          "messages": [
            {
              "role": "system"
            }
          ]
        }
      }'

Then paste one of the prompts above: Claude Code finds Grok 4.20 on Spiral, shows the price and runs it.

Grok 4.20 providers and prices

Spiral runs Grok 4.20 through OpenRouter. Open a card for today's price, with every fee included.

OpenRouterLLM
SpaceXAI: Grok 4.20
x-ai/grok-4.20

Settings you can ask for

Ask Claude Code for any of these in your prompt.

SettingAllowed valuesDefaultDescription
Seeda whole number
Top pa number up to 1
Max tokens1 to 1800000

The most output tokens to generate.

Temperature0 to 2

Every API field of SpaceXAI: Grok 4.20 is on its tool page: open its card above.

What can you do with Grok 4.20 in Claude Code?

Grok 4.20 can help turn project code and requirements into a debugging plan, a patch, or a migration review. Give it the relevant files, acceptance criteria, and required checks so the result is easy to assess.

Compare technical proposals

Ask Grok 4.20 to compare competing proposals against explicit requirements, identify tradeoffs, and cite the passages supporting each conclusion.

Investigate a software defect

Use Grok 4.20 to trace a reported failure through relevant code, propose a cause, and design a regression test that would distinguish it from other explanations.

Extract document records

Have Grok 4.20 turn supplied invoice or order text into JSON records, preserving source values and marking missing information explicitly.

Synthesize research evidence

Give Grok 4.20 dated excerpts and ask for findings, disagreements, and unanswered questions, with a source reference beside each finding.

Check quantitative reasoning

Ask Grok 4.20 to examine a calculation, identify assumptions, check units, and explain which inputs would change the conclusion.

Translate support content

Use Grok 4.20 to draft translations in English, Spanish, Chinese, Arabic, or Hindi while preserving product names and a supplied glossary.

Draft evidence-based reports

Have Grok 4.20 turn supplied operational updates into a report with decisions, unresolved issues, and supporting evidence.

Review a migration plan

Ask Grok 4.20 to check dependencies, sequence work, identify rollback gaps, and propose acceptance checks against the stated requirements.

Questions about Grok 4.20 in Claude Code

Can Claude Code use Grok 4.20?

Yes. Add Spiral to Claude Code as an MCP server, and Claude Code can find and run Grok 4.20 like any built-in tool.

How much does Grok 4.20 cost in Claude Code?

Up to $0.01337 per run at the default inputs. $1.625 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 Claude Code plan is separate.

Do I need an OpenRouter account to use Grok 4.20 in Claude Code?

No. Spiral runs Grok 4.20 on its own OpenRouter account. You sign in to Spiral once and never paste a key.

How long does it take to set up Grok 4.20 in Claude Code?

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

Does Claude Code wait while Grok 4.20 runs?

Grok 4.20 runs in the background. Claude Code starts the run, checks on it and gives you the text, json when it's ready.

What is Grok 4.20 useful for?

Grok 4.20 supports research, writing, coding, translation, and reasoning over supplied material. Good starting tasks include comparing technical proposals, explaining unfamiliar code, and turning document text into structured records.

How much context can Grok 4.20 handle?

The exact catalog listing reports 2 million tokens, while xAI's model page lists 1 million. These published limits disagree, so use 1 million tokens as a conservative planning ceiling; neither figure promises an answer of that length.

Which languages does Grok 4.20 support?

xAI names English, Chinese, Arabic, Hindi, and Spanish, plus several dozen others. Capability varies by language, so review terminology and meaning before publishing a translation.

How should I prompt Grok 4.20?

State the task, constraints, and required output directly. For example, ask for a comparison of three proposals using named criteria, a word limit, and a separate list of missing evidence.

Which settings are worth testing first?

Compare reasoning enabled and disabled on representative tasks. For sampling, xAI recommends temperature 0.7 and top-p 0.95; use those as a starting point.

Get started

Start using Grok 4.20 in Claude Code

One Spiral account for every tool your agent needs. No subscription; you pay per run.

Paste this into Claude Code:

Set up Claude Code

set up https://spiral.ai/SKILL.md
Are you an AI agent? See setup options