How to Use Qwen3 14B in Claude Code
Qwen3 14B suits coding, math, and multilingual text work. It combines thinking for harder problems with a direct-answer mode for everyday conversation in one model.
Prompts to try with Qwen3 14B in Claude Code
Copy an example into Claude Code, or ask in your own words.
Use Qwen3 14B in thinking mode to inspect the failing test and relevant implementation in this project. Propose the smallest fix, add a regression test, and explain the defect in three sentences.
Use Qwen3 14B to review this project's CSV parser for quoted commas, embedded newlines, and empty cells. Report each suspected defect with a minimal input that would reproduce it.
Use Qwen3 14B in direct-answer mode to rewrite the README's installation explanation for a new contributor. Preserve every command and keep the explanation under 250 words.
Use Qwen3 14B to translate the English interface strings in this project into Spanish. Preserve interpolation tokens and return a list of phrases whose meaning needs clarification.
Use Qwen3 14B in thinking mode to review the discount calculation and its tests. Check rounding, zero quantities, and overlapping discounts, then propose tests for uncovered behavior.
How to set up Qwen3 14B 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 14B
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/qwen3-14b", "input": { "messages": [ { "role": "system" } ] } }'
Then paste one of the prompts above: Claude Code finds Qwen3 14B on Spiral, shows the price and runs it.
Qwen3 14B providers and prices
Spiral runs Qwen3 14B through OpenRouter. Open a card for today's price, with every fee included.
Up to $0.001285 per run at the default inputs. $0.156 per 1M input tokens, $0.312 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 16384 | 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 14B is on its tool page: open its card above.
What can you do with Qwen3 14B in Claude Code?
Turn project code and failure reports into proposed fixes, tests, and explanations with Qwen3 14B. Use thinking for debugging that requires several reasoning steps and direct answers for routine documentation.
Debug a failing function
Use Qwen3 14B to examine code and an error trace, propose a correction, and identify a test that would catch the defect.
Check a mathematical solution
Give Qwen3 14B the problem and proposed solution. Ask it to check assumptions, recompute the result, and explain any discrepancy.
Translate customer-facing text
Ask Qwen3 14B to translate a support reply while preserving product names, technical terms, and the requested tone.
Extract structured records
Turn supplied messages into records with a defined structure. Specify which values to extract and how to represent missing information.
Practice a difficult conversation
Have Qwen3 14B play a customer or interviewer, respond one turn at a time, and provide feedback after the exercise.
Draft short creative scenes
Give Qwen3 14B a character, setting, and word limit. Request a scene with concrete constraints on dialogue, tone, and ending.
Answer questions from supplied passages
Use Qwen3 14B to answer a focused question from text provided in the prompt. Require supporting excerpts and an explicit response when evidence is missing.
More on Qwen3 14B and Claude Code
Overview
Qwen3 14B in other agents
Questions about Qwen3 14B in Claude Code
Can Claude Code use Qwen3 14B?
Yes. Add Spiral to Claude Code as an MCP server, and Claude Code can find and run Qwen3 14B like any built-in tool.
How much does Qwen3 14B cost in Claude Code?
Up to $0.001285 per run at the default inputs. $0.156 per 1M input tokens, $0.312 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 14B in Claude Code?
No. Spiral runs Qwen3 14B 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 14B in Claude Code?
About two minutes: add the hosted server, then sign in.
Does Claude Code wait while Qwen3 14B runs?
Qwen3 14B 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 14B useful for?
Use it for coding, mathematical reasoning, multilingual text, and conversations with detailed instructions. It also supports creative writing and role-play.
When should I use thinking mode?
Choose thinking for problems with several reasoning steps, such as debugging or mathematical analysis. Use non-thinking mode for straightforward conversation and routine text work.
How much context can Qwen3 14B handle?
The model card specifies 32,768 tokens natively and 131,072 with YaRN extension. Treat the larger number as conditional on that extension being enabled.
Can Qwen3 14B work across languages?
Qwen reports support for 119 languages and dialects. Specify the target language, intended audience, and terms that should stay unchanged when requesting a translation.
What can cause repetitive answers?
Qwen warns that greedy decoding in thinking mode can cause performance degradation and endless repetition. Its recommended thinking settings use temperature 0.6, top-p 0.95, top-k 20, and min-p 0.
Start using Qwen3 14B 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


