How to Use Ling 3.1 Flash in Codex
Ling 3.1 Flash handles coding, document analysis, and tasks that use tools. Its hybrid reasoning design supports both deliberate problem solving and direct answers, with a model card advertising 1 million tokens of context.
Prompts to try with Ling 3.1 Flash in Codex
Copy an example into Codex, or ask in your own words.
Use Ling 3.1 Flash 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.
Use Ling 3.1 Flash to review the current diff for correctness. Report each issue with its file location, triggering input, and a test that would expose it.
Use Ling 3.1 Flash to compare the implementation with the requirements in this repository. List unmet requirements with code references, then propose the order to address them.
Use Ling 3.1 Flash to inspect this project's retry logic. Identify where retries could repeat a side effect, add a focused regression test, and propose a fix.
Use Ling 3.1 Flash to read the project notes and recent changes. Summarize completed work, unresolved blockers, and the next three actions with a verification step for each.
How to set up Ling 3.1 Flash 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 Ling 3.1 Flash
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": "inclusionai/ling-3.1-flash", "input": { "messages": [ { "role": "system" } ] } }'
Then paste one of the prompts above: Codex finds Ling 3.1 Flash on Spiral, shows the price and runs it.
Ling 3.1 Flash providers and prices
Spiral runs Ling 3.1 Flash through OpenRouter. Open a card for today's price, with every fee included.
Up to $0.061346 per run at the default inputs. $0.078 per 1M input tokens, $0.234 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 262144 | 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 inclusionAI: Ling 3.1 Flash is on its tool page: open its card above.
What can you do with Ling 3.1 Flash in Codex?
Use Ling 3.1 Flash to investigate bugs, review changes, and turn requirements into code and tests. Give it relevant project files and a concrete check for completion.
Investigate a failing test
Use Ling 3.1 Flash to trace a failure through relevant code, propose a focused fix, and draft a regression test. Check the proposal by running the test.
Review a code change
Ask Ling 3.1 Flash to examine a diff against stated requirements and identify concrete failure conditions. Request a reproduction for each reported issue.
Compare lengthy specifications
Give Ling 3.1 Flash two specifications as text and request a table of changed requirements, conflicting statements, and unanswered questions, with section references.
Prepare a decision brief
Use Ling 3.1 Flash to compare proposals against explicit constraints. Ask it to connect each recommendation to supplied evidence and identify missing information.
Resume a lengthy task
Give Ling 3.1 Flash the task history, completed work, and unresolved blockers. Request the next actions with a clear completion condition for each.
Investigate an operational incident
Use Ling 3.1 Flash to compare timestamped logs, incident notes, and tool results. Ask for a timeline and separate observed events from possible causes.
Draft a recurring change digest
Ask Ling 3.1 Flash to compare newly collected text with the previous reporting period. Request only substantive changes, supporting excerpts, and follow-up questions.
Extract records for review
Ask Ling 3.1 Flash to turn supplied text into JSON records with supporting excerpts. Validate the JSON and check required values before accepting the records.
More on Ling 3.1 Flash and Codex
Overview
Ling 3.1 Flash in other agents
Questions about Ling 3.1 Flash in Codex
Can Codex use Ling 3.1 Flash?
Yes. Add Spiral to Codex as an MCP server, and Codex can find and run Ling 3.1 Flash like any built-in tool.
How much does Ling 3.1 Flash cost in Codex?
Up to $0.000722 per run at the default inputs. $0.0585 per 1M input tokens, $0.1755 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 Ling 3.1 Flash in Codex?
No. Spiral runs Ling 3.1 Flash on its own OpenRouter account. You sign in to Spiral once and never paste a key.
How long does it take to set up Ling 3.1 Flash in Codex?
About two minutes: add the hosted server, then sign in.
Does Codex wait while Ling 3.1 Flash runs?
Ling 3.1 Flash runs in the background. Codex starts the run, checks on it and gives you the text, json when it's ready.
What is Ling 3.1 Flash good at?
Ling 3.1 Flash is designed for coding, analysis involving several steps, and tasks that use tools. Practical applications include investigating code, comparing long documents, and continuing work from an extended task history.
Can Ling 3.1 Flash help with software development?
Yes. Coding is a documented focus, making bug investigation, code review, and test drafting useful tasks to try. Give it the relevant code, the expected behavior, and a way to check the result.
How much text can Ling 3.1 Flash handle?
The maker advertises 1 million tokens of context; SGLang clarifies that native context is 262,144 tokens, extended to 1,048,576 with YaRN. OpenRouter separately lists up to 262,144 completion tokens. Keep the prompt and requested response within the configured context limit.
Does Ling 3.1 Flash guarantee valid JSON?
No. OpenRouter documents that JSON output is not enforced. Request an exact structure, then parse and validate the response before using it in an automated workflow.
Which sampling settings does the maker recommend?
The model card recommends temperature 1.0, top_p 0.95, and top_k 20, with thinking enabled for better performance. Use those values as a starting point when evaluating your task.
Start using Ling 3.1 Flash 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


