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

Claude Code · LLMReleased 2026-08-14

How to Use Qwen3.8 27B in Claude Code

Qwen3.8 27B suits code changes, research synthesis, and tasks that require several rounds of decisions. Its 27-billion-parameter dense design combines adjustable thinking with a native 262,144-token context.

Prompts to try with Qwen3.8 27B in Claude Code

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

Use Qwen3.8 27B to investigate the failing tests in this repository. Trace the failure through the relevant code, make the smallest justified fix, and run the affected tests.

Qwen3.8 27BClaude Code
Try asking
  • Use Qwen3.8 27B to review the current diff for correctness bugs. For each finding, identify the file, triggering input, and resulting failure; omit style preferences.

  • Use Qwen3.8 27B to turn the requirements in this project's README into an implementation plan. Identify dependencies and missing acceptance criteria before proposing edits.

  • Use Qwen3.8 27B to examine this repository's request validation. Find one concrete case where invalid input reaches business logic, then propose a fix and a regression test.

  • Use Qwen3.8 27B to explain the authentication flow in this project. Follow the actual call path and cite the relevant files; mark any behavior you cannot establish from the code.

How to set up Qwen3.8 27B 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 Qwen3.8 27B

    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": "qwen/qwen3.8-27b",
        "input": {
          "messages": [
            {
              "role": "system"
            }
          ]
        }
      }'

Then paste one of the prompts above: Claude Code finds Qwen3.8 27B on Spiral, shows the price and runs it.

Qwen3.8 27B providers and prices

Spiral runs Qwen3.8 27B through OpenRouter. Open a card for today's price, with every fee included.

OpenRouterLLM
Qwen: Qwen3.8 27B
qwen/qwen3.8-27b

Settings you can ask for

Ask Claude Code for any of these in your prompt.

SettingAllowed valuesDefaultDescription
Seeda whole number
Min p0 to 1
Top ka whole number from 0
Top p0 to 1
Logprobson or off
Max tokens1 to 235929

The most output tokens to generate.

Temperature0 to 2
Top logprobs0 to 20
Presence penalty-2 to 2
Reasoning effortnone, low, medium, or xhighxhigh

How much the model reasons.

Frequency penalty-2 to 2
Repetition penalty0 to 2

Every API field of Qwen: Qwen3.8 27B is on its tool page: open its card above.

What can you do with Qwen3.8 27B in Claude Code?

Turn repository context into targeted fixes, reviews, and implementation plans with Qwen3.8 27B. Give the agent a reproducible failure or acceptance criteria so it has a concrete result to check.

Bug diagnosis

Give Qwen3.8 27B the failing behavior, relevant code, and error messages. Request a likely cause, a minimal correction, and a regression check.

Code review

Ask Qwen3.8 27B to review a proposed change for concrete failure paths. Require each finding to include the triggering condition and affected code.

Research comparison

Supply research excerpts and ask Qwen3.8 27B to compare claims, methods, and disagreements, with a source reference for each conclusion.

Requirements analysis

Have Qwen3.8 27B turn a product brief into acceptance criteria, dependencies, and unresolved questions.

Document synthesis

Use Qwen3.8 27B to combine supplied reports into a decision brief that separates evidence, assumptions, and missing information.

Incident review

Give Qwen3.8 27B a text timeline and logs. Request an evidence-based sequence of events and follow-up actions.

Recurring change reports

Ask Qwen3.8 27B to compare current and previous updates, identify material changes, and list decisions that need attention.

Technical explanations

Have Qwen3.8 27B explain supplied code or technical notes for a specific audience, using examples drawn from the source material.

Questions about Qwen3.8 27B in Claude Code

Can Claude Code use Qwen3.8 27B?

Yes. Add Spiral to Claude Code as an MCP server, and Claude Code can find and run Qwen3.8 27B like any built-in tool.

How much does Qwen3.8 27B cost in Claude Code?

Up to $0.009992 per run at the default inputs. $0.195 per 1M input tokens, $2.4375 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 Qwen3.8 27B in Claude Code?

No. Spiral runs Qwen3.8 27B on its own OpenRouter account. You sign in to Spiral once and never paste a key.

How long does it take to set up Qwen3.8 27B in Claude Code?

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

Does Claude Code wait while Qwen3.8 27B runs?

Qwen3.8 27B runs in the background. Claude Code starts the run, checks on it and gives you the text, json when it's ready.

What is Qwen3.8 27B useful for?

Consider it for code changes, research synthesis, and work that requires several decisions. Qwen highlights improvements in coding, professional work, research, and extended agent tasks.

How much context does Qwen3.8 27B support?

Its native context is 262,144 tokens. The documented extension to 1,000,000 tokens requires additional configuration, so treat that as an extended capability rather than the native limit.

When should I use thinking?

Use thinking for debugging, comparing evidence, or planning work with dependencies. Try direct answers for straightforward extraction and rewriting; Qwen3.8 27B supports both approaches.

What sampling settings should I start with?

The shipped configuration provides a starting point: temperature 1.0, top-p 0.95, and top-k 20. Compare results on representative tasks before changing several settings at once.

What weakness should I watch for when reducing repetition?

Aggressive repetition suppression can introduce mixed-language output and reduce quality. If that happens, reduce the repetition penalty before rewriting the whole prompt.

Get started

Start using Qwen3.8 27B 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