How to Use Qwen3.8 27B (batch) in opencode
Qwen3.8 27B suits code repair and multi-step office work. Compared with Qwen3.6 27B, it improves coding and office productivity across text and visual tasks.
Prompts to try with Qwen3.8 27B (batch) in opencode
Copy an example into opencode, or ask in your own words.
Use Qwen3.8 27B to inspect the failing tests in this project, trace the cause, and propose the smallest justified patch. Run the relevant tests and report unresolved failures.
Use Qwen3.8 27B to review the current diff against the project's acceptance criteria. List concrete behavior changes and missing regression checks.
Use Qwen3.8 27B to turn the requirements in this project's issue notes into an implementation plan. Identify affected files, dependencies, and a verification step for each change.
Use Qwen3.8 27B to refactor duplicated validation in this project while preserving behavior. Start with temperature 1.0, top_p 0.95, and top_k 20; verify the result with relevant tests.
Use Qwen3.8 27B to compare the project's public interfaces with its documentation. Draft corrections for confirmed mismatches and identify claims that need further checking.
How to set up Qwen3.8 27B (batch) in opencode
Pick the way that suits you. Each takes about two minutes.
Use Spiral's MCP server
- Add the server
Leave out --global to add Spiral to the current project only. No header and no client ID or secret: opencode registers itself with Spiral.
opencode mcp add spiral --global --url https://api.spiral.ai/mcp
- Sign in
Your browser opens. Sign in to Spiral, pick the organization and click Authorize, then return to the terminal. No API key. Inside opencode, /mcps starts the same sign-in.
opencode mcp auth spiral
- Check it
Spiral shows as connected.
opencode mcp list
Use the Spiral skill
- Send opencode 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 Qwen3.8 27B (batch)
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/qwen3.8-27b:batch", "input": { "messages": [ { "role": "system" } ] } }'
Then paste one of the prompts above: opencode finds Qwen3.8 27B (batch) on Spiral, shows the price and runs it.
Qwen3.8 27B (batch) providers and prices
Spiral runs Qwen3.8 27B (batch) through OpenRouter. Open a card for today's price, with every fee included.
Up to $0.010658 per run at the default inputs. $0.208 per 1M input tokens, $2.60 per 1M output tokens.
Settings you can ask for
Ask opencode 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 235929 | The most output tokens to generate. On a reasoning model the reasoning tokens count toward it: a cap too low for the effort can end with no answer (finish_reason length). | |
| Temperature | 0 to 2 | ||
| Presence penalty | -2 to 2 | ||
| Reasoning effort | none, low, medium, or xhigh | xhigh | How much the model reasons. |
| Frequency penalty | -2 to 2 | ||
| Repetition penalty | 0 to 2 |
Every API field of Qwen: Qwen3.8 27B (batch) is on its tool page: open its card above.
What can you do with Qwen3.8 27B (batch) in opencode?
Qwen3.8 27B can help turn project requirements into code changes and verification plans. Give it a bounded task with explicit acceptance criteria.
Repair a reproducible bug
Use Qwen3.8 27B to trace a failing example, propose a focused patch, and explain how to check the fix.
Write tests from requirements
Ask Qwen3.8 27B to turn acceptance criteria into tests covering expected behavior, invalid input, and boundary conditions.
Plan a code migration
Give Qwen3.8 27B the current behavior and target requirements, then request an ordered migration plan with checks at each stage.
Review a proposed refactor
Have Qwen3.8 27B compare the original and revised code, identify behavior changes, and suggest regression checks.
Compare document revisions
Ask Qwen3.8 27B to identify changed requirements, conflicting statements, and decisions that need an owner's review.
Turn meeting notes into actions
Use Qwen3.8 27B to draft an action register, preserving stated owners and deadlines while marking missing details.
Draft a project status brief
Give Qwen3.8 27B dated updates and request a brief covering completed work, blockers, dependencies, and next decisions.
More on Qwen3.8 27B (batch) and opencode
Overview
Qwen3.8 27B (batch) in other agents
Questions about Qwen3.8 27B (batch) in opencode
Can opencode use Qwen3.8 27B (batch)?
Yes. Add Spiral to opencode as an MCP server, and opencode can find and run Qwen3.8 27B (batch) like any built-in tool.
How much does Qwen3.8 27B (batch) cost in opencode?
Up to $0.010658 per run at the default inputs. $0.208 per 1M input tokens, $2.60 per 1M output tokens. You pay from your Spiral balance as you go, with no subscription, and a failed run is not charged. Your opencode plan is separate.
Do I need an OpenRouter account to use Qwen3.8 27B (batch) in opencode?
No. Spiral runs Qwen3.8 27B (batch) on its own OpenRouter account. You sign in to Spiral once and never paste a key.
How long does it take to set up Qwen3.8 27B (batch) in opencode?
About two minutes: add the server, then sign in, then check it.
Does opencode wait while Qwen3.8 27B (batch) runs?
Qwen3.8 27B (batch) runs in the background. opencode starts the run, checks on it and gives you the text, json when it's ready.
What is Qwen3.8 27B useful for?
Start with code repair, implementation planning, and office tasks that require several steps. Qwen reports improvements over Qwen3.6 27B in coding and office productivity.
How much context can Qwen3.8 27B handle?
Its native context is 262,144 tokens, with extension to 1,000,000 tokens. Treat the larger figure as an extended configuration rather than the native limit.
How long can a Qwen3.8 27B answer be?
Qwen recommends allowing up to 131,072 final-response tokens when using an extended 1M context and separate reasoning and answer limits. That recommendation is conditional, rather than a universal output allowance.
Does Qwen3.8 27B have to reason before answering?
It supports both adaptive thinking and direct responses without thinking. Try direct responses for straightforward extraction and thinking for tasks that require several decisions.
Which sampling settings should I start with?
For thinking tasks, start with the released configuration: temperature 1.0, top_p 0.95, and top_k 20. Compare results on representative tasks before changing those settings.
Start using Qwen3.8 27B (batch) in opencode
One Spiral account for every tool your agent needs. No subscription; you pay per run.
Paste this into opencode:
set up https://spiral.ai/SKILL.md


