How to Use Qwen3.5-9B in Claude Code
Use Qwen3.5-9B for coding, document questions, and multilingual drafts. Artificial Analysis's March 2026 evaluation gave it an Intelligence Index score of 32, compared with 27 for Qwen3.5-4B.
Prompts to try with Qwen3.5-9B in Claude Code
Copy an example into Claude Code, or ask in your own words.
Use Qwen3.5-9B to inspect the failing tests and relevant source files. Find the cause, make the smallest justified patch, and run the affected tests.
Use Qwen3.5-9B to review the current diff for behavior changes. Report only issues supported by the code, with file locations and a reproducing example.
Use Qwen3.5-9B to add tests for the parser in this project. Cover empty input, malformed input, Unicode text, and boundary values; derive expected results from the documented contract.
Use Qwen3.5-9B in thinking mode with temperature=0.6, top_p=0.95, top_k=20, and min_p=0 to implement the requested form validation. Follow the project's existing conventions and verify each validation rule.
Use Qwen3.5-9B to compare the implementation with the project's API specification. List mismatches with evidence, then propose a patch for the clearest mismatch.
How to set up Qwen3.5-9B in Claude Code
Pick the way that suits you. Each takes about two minutes.
Use Spiral's MCP server
- 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
- 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
- Send Claude Code 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 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 Qwen3.5-9B
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/qwen3.5-9b", "input": { "messages": [ { "role": "system" } ] } }'
Then paste one of the prompts above: Claude Code finds Qwen3.5-9B on Spiral, shows the price and runs it.
Qwen3.5-9B providers and prices
Spiral runs Qwen3.5-9B through OpenRouter. Open a card for today's price, with every fee included.
Up to $0.001339 per run at the default inputs. $0.221 per 1M input tokens, $0.325 per 1M output tokens.
Settings you can ask for
Ask Claude Code 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 Qwen: Qwen3.5-9B is on its tool page: open its card above.
What can you do with Qwen3.5-9B in Claude Code?
Qwen3.5-9B can help turn project code and error reports into focused patches, tests, and explanations. Give it a bounded task with a checkable result.
Fix a reproducible bug
Ask Qwen3.5-9B to inspect the failing behavior, propose a small patch, and identify a regression test. Supply the relevant code and error output.
Turn requirements into tests
Give Qwen3.5-9B a function contract and ask for normal cases, boundary cases, and invalid inputs with expected results.
Answer questions about long documents
Ask Qwen3.5-9B to compare supplied policies or specifications and cite the passages behind each conclusion.
Draft localized support replies
Ask Qwen3.5-9B to rewrite a supplied response for a named language and locale while preserving product terms and policy details.
Extract records from text
Give Qwen3.5-9B messages or notes and request structured records with evidence excerpts and explicit missing values.
Compare operational plans
Ask Qwen3.5-9B to compare proposed schedules or rollout plans against explicit constraints, then explain conflicts and unresolved assumptions.
More on Qwen3.5-9B and Claude Code
Overview
Qwen3.5-9B in other agents
More Claude Code guides
Questions about Qwen3.5-9B in Claude Code
Can Claude Code use Qwen3.5-9B?
Yes. Add Spiral to Claude Code as an MCP server, and Claude Code can find and run Qwen3.5-9B like any built-in tool.
How much does Qwen3.5-9B cost in Claude Code?
Up to $0.000172 per run at the default inputs. $0.0286 per 1M input tokens, $0.0416 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 Qwen3.5-9B in Claude Code?
No. Spiral runs Qwen3.5-9B on its own OpenRouter account. You sign in to Spiral once and never paste a key.
How long does it take to set up Qwen3.5-9B in Claude Code?
About two minutes: add the hosted server, then sign in.
Does Claude Code wait while Qwen3.5-9B runs?
Qwen3.5-9B runs in the background. Claude Code starts the run, checks on it and gives you the text, json when it's ready.
What should I use Qwen3.5-9B for?
Try bounded coding tasks, questions about supplied documents, and multilingual drafts. Give it the relevant source material and a clear definition of the finished answer.
How much context can Qwen3.5-9B handle?
Qwen documents a native context window of 262,144 tokens. Its documented extension to 1,010,000 tokens is separate from the native limit.
How long should I let Qwen3.5-9B generate?
Qwen recommends a 32,768-token output budget for most queries. That budget is guidance, not a required answer length or a stated maximum.
Can Qwen3.5-9B work across languages?
Alibaba reports support for 201 languages and dialects across Qwen3.5. For translation or localization, specify the target locale, audience, and terms to preserve.
How does Qwen3.5-9B compare with Qwen3.5-4B?
Artificial Analysis's March 2026 evaluation scored Qwen3.5-9B at 32 and Qwen3.5-4B at 27 on its Intelligence Index. Treat that as an overall benchmark comparison and test both on your actual task.
Start using Qwen3.5-9B 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 https://spiral.ai/SKILL.md


