How to Use Qwen3 30B A3B in Codex
Qwen3 30B A3B handles coding, math, translation, and tasks that use external tools. It combines thinking and direct-answer modes in one model, with 30.5 billion total parameters and 3.3 billion activated.
Prompts to try with Qwen3 30B A3B in Codex
Copy an example into Codex, or ask in your own words.
Use Qwen3 30B A3B in thinking mode to inspect the failing test and relevant source files. Propose the smallest fix and a regression test.
Use Qwen3 30B A3B to review the changed functions for empty inputs, off-by-one errors, and incorrect assumptions. Report each issue with a concrete example.
Use Qwen3 30B A3B in direct-answer mode to rewrite the README introduction for a first-time user. Preserve every command and technical claim.
Use Qwen3 30B A3B to translate the English interface strings into Spanish. Preserve interpolation tokens and return the same key structure.
Use Qwen3 30B A3B in thinking mode to implement the validation rules in the project specification. Include tests for each boundary condition.
How to set up Qwen3 30B A3B in Codex
Pick the way that suits you. Each takes about two minutes.
Use Spiral's MCP server
- Add the hosted server
codex mcp add spiral --url https://api.spiral.ai/mcp
- Sign in
Codex opens the browser: sign in to Spiral and click Authorize. No API key. codex mcp list then shows Spiral.
codex mcp login spiral
Use the Spiral skill
- Send Codex 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 30B A3B
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/qwen3-30b-a3b", "input": { "messages": [ { "role": "system" } ] } }'
Then paste one of the prompts above: Codex finds Qwen3 30B A3B on Spiral, shows the price and runs it.
Qwen3 30B A3B providers and prices
Spiral runs Qwen3 30B A3B through OpenRouter. Open a card for today's price, with every fee included.
Up to $0.002669 per run at the default inputs. $0.156 per 1M input tokens, $0.65 per 1M output tokens.
Settings you can ask for
Ask Codex 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 | ||
| Max tokens | 1 to 16384 | The most output tokens to generate. | |
| Temperature | 0 to 2 | ||
| Presence penalty | -2 to 2 | ||
| Frequency penalty | -2 to 2 | ||
| Repetition penalty | 0 to 2 |
Every API field of Qwen: Qwen3 30B A3B is on its tool page: open its card above.
What can you do with Qwen3 30B A3B in Codex?
Turn code and requirements into proposed fixes, utility functions, and tests with Qwen3 30B A3B. Use thinking mode for debugging and logic, and direct answers for straightforward edits.
Debug a failing function
Give Qwen3 30B A3B the code, error, and expected behavior. Ask for a proposed fix and a test that checks the failure.
Check a math solution
Ask Qwen3 30B A3B to solve a problem in thinking mode, then compare the result with your attempted solution.
Translate customer messages
Ask Qwen3 30B A3B to translate a message while preserving names, product terms, and the requested level of formality.
Draft support replies
Supply the customer’s question and approved policy text. Ask Qwen3 30B A3B for a concise reply that identifies any missing information.
Revise a rough draft
Give Qwen3 30B A3B a draft, audience, and length target. Ask for clearer wording while preserving the original claims.
Practice a difficult conversation
Have Qwen3 30B A3B play a customer, interviewer, or colleague. Continue across several turns, then request feedback on your responses.
Write small utility functions
Ask Qwen3 30B A3B to turn a precise specification into code, including examples for empty inputs and boundary conditions.
Answer questions using tools
Have Qwen3 30B A3B use available lookup or calculation tools, then write an answer grounded in their returned results.
More on Qwen3 30B A3B and Codex
Overview
Qwen3 30B A3B in other agents
Questions about Qwen3 30B A3B in Codex
Can Codex use Qwen3 30B A3B?
Yes. Add Spiral to Codex as an MCP server, and Codex can find and run Qwen3 30B A3B like any built-in tool.
How much does Qwen3 30B A3B cost in Codex?
Up to $0.002669 per run at the default inputs. $0.156 per 1M input tokens, $0.65 per 1M output tokens. You pay from your Spiral balance as you go, with no subscription, and a failed run is not charged. Your Codex plan is separate.
Do I need an OpenRouter account to use Qwen3 30B A3B in Codex?
No. Spiral runs Qwen3 30B A3B 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 30B A3B in Codex?
About two minutes: add the hosted server, then sign in.
Does Codex wait while Qwen3 30B A3B runs?
Qwen3 30B A3B runs in the background. Codex starts the run, checks on it and gives you the text, json when it's ready.
What is Qwen3 30B A3B good at?
Use it for coding, math, translation, writing, and tasks that involve external tools. Thinking mode suits difficult reasoning; direct-answer mode suits simpler requests.
What does A3B mean?
It refers to the activated parameter count. Qwen3 30B A3B has 30.5 billion parameters in total, with 3.3 billion activated.
How much text can Qwen3 30B A3B handle?
The maker documents a native context of 32,768 tokens and extension to 131,072 tokens with YaRN. The larger figure requires context extension.
Which languages does Qwen3 30B A3B support?
Qwen reports support for 119 languages and dialects. Its published list includes English, Spanish, Chinese, Arabic, Japanese, and Korean.
When should I use thinking mode?
Choose thinking mode for difficult code, math, or logic problems. Choose direct answers for straightforward dialogue and rewriting.
Start using Qwen3 30B A3B in Codex
One Spiral account for every tool your agent needs. No subscription; you pay per run.
Paste this into Codex:
set up https://spiral.ai/SKILL.md


