How to Use Qwen3.5-122B-A10B in Codex
Qwen3.5-122B-A10B suits coding, tool-driven tasks, and analysis of long texts across languages. Its mixture-of-experts design has 122 billion total parameters, with 10 billion activated.
Prompts to try with Qwen3.5-122B-A10B in Codex
Copy an example into Codex, or ask in your own words.
Use Qwen3.5-122B-A10B to investigate the failing tests in this project. Trace the failure through the relevant files, make the smallest justified fix, and run the affected tests. Report the cause and verification results.
Use Qwen3.5-122B-A10B to review this project's authentication flow. Follow the code from entry point to session creation, cite file paths, and list behavior that needs clarification before proposing changes.
Use Qwen3.5-122B-A10B in thinking mode with temperature 0.6, top_p 0.95, top_k 20, and min_p 0.0 to implement pagination in the existing list endpoint. Follow the project's conventions and verify empty, first, and last pages.
Use Qwen3.5-122B-A10B to compare this project's implementation with its requirements documents. Return a table containing each requirement, the supporting code path, any gap, and a suggested verification step.
Use Qwen3.5-122B-A10B to review the English and Spanish UI strings in this project. Preserve interpolation tokens and product names, flag meaning changes, and propose corrections with their file paths.
How to set up Qwen3.5-122B-A10B 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 Qwen3.5-122B-A10B
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/qwen3.5-122b-a10b", "input": { "messages": [ { "role": "system" } ] } }'
Then paste one of the prompts above: Codex finds Qwen3.5-122B-A10B on Spiral, shows the price and runs it.
Qwen3.5-122B-A10B providers and prices
Spiral runs Qwen3.5-122B-A10B through OpenRouter. Open a card for today's price, with every fee included.
Up to $0.011088 per run at the default inputs. $0.338 per 1M input tokens, $2.704 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 | ||
| Logprobs | on or off | ||
| Max tokens | 1 to 65536 | 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: Qwen3.5-122B-A10B is on its tool page: open its card above.
What can you do with Qwen3.5-122B-A10B in Codex?
Turn a codebase question into an investigation, patch, and verification plan with Qwen3.5-122B-A10B. Ask the agent to connect relevant project files and tool results, then explain the change against explicit acceptance criteria.
Trace a bug across a codebase
Ask Qwen3.5-122B-A10B to connect a failing test to the relevant functions, propose a focused patch, and explain the evidence.
Plan a code migration
Use Qwen3.5-122B-A10B to compare current code with supplied migration guidance and list required changes, dependencies, and checks.
Compare lengthy requirements
Give Qwen3.5-122B-A10B labeled requirements and design notes. Request a table of agreements, conflicts, and unanswered questions with passage references.
Prepare multilingual support responses
Ask Qwen3.5-122B-A10B to draft replies from supplied support guidance, preserving product names and following a regional terminology list.
Build a source-based research brief
Have Qwen3.5-122B-A10B compare retrieved passages, cite their source labels, and separate supported conclusions from open questions.
Review recurring operational updates
Use Qwen3.5-122B-A10B to turn retrieved status updates into a concise digest of changes, blockers, and explicitly stated next actions.
Extract structured records from text
Ask Qwen3.5-122B-A10B to organize supplied text into a defined JSON structure, with source references and explicit missing values.
Coordinate a tool-assisted investigation
Give Qwen3.5-122B-A10B a question and a defined set of retrieval tools. Request evidence collection followed by a concise answer citing the returned results.
More on Qwen3.5-122B-A10B and Codex
Overview
Qwen3.5-122B-A10B in other agents
Questions about Qwen3.5-122B-A10B in Codex
Can Codex use Qwen3.5-122B-A10B?
Yes. Add Spiral to Codex as an MCP server, and Codex can find and run Qwen3.5-122B-A10B like any built-in tool.
How much does Qwen3.5-122B-A10B cost in Codex?
Up to $0.011088 per run at the default inputs. $0.338 per 1M input tokens, $2.704 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 Qwen3.5-122B-A10B in Codex?
No. Spiral runs Qwen3.5-122B-A10B on its own OpenRouter account. You sign in to Spiral once and never paste a key.
How long does it take to set up Qwen3.5-122B-A10B in Codex?
About two minutes: add the hosted server, then sign in.
Does Codex wait while Qwen3.5-122B-A10B runs?
Qwen3.5-122B-A10B runs in the background. Codex starts the run, checks on it and gives you the text, json when it's ready.
What is Qwen3.5-122B-A10B useful for?
Consider it for codebase analysis, tool-driven workflows, and long-text tasks across languages. Suggested projects include tracing a bug, comparing requirements, or preparing a multilingual support brief.
What does 122B-A10B mean?
It describes a mixture-of-experts model with 122 billion total parameters and 10 billion activated. The two numbers describe total size and active computation.
How much text can Qwen3.5-122B-A10B handle?
Its native context is 262,144 tokens, counting input and output together. Qwen documents extension to 1,010,000 tokens with context scaling, which can affect shorter-text performance.
Which languages does Qwen3.5-122B-A10B support?
Qwen reports coverage of 201 languages and dialects. Specify the target language, regional wording, and terms to preserve when requesting a translation.
Which settings should I try for coding?
For precise coding in thinking mode, Qwen recommends temperature 0.6, top-p 0.95, top-k 20, and min-p zero. Treat these as starting values and assess the result against the task’s acceptance criteria.
Start using Qwen3.5-122B-A10B 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


