How to Use ERNIE-Light-DT-1.0 in Muse Code
ERNIE-Light-DT-1.0 is built for long-document questions, summaries, coding, and reasoning. Its 1,024,000-token context window supports large reference sets, while optional thinking lets you match the reasoning effort to the task.
Prompts to try with ERNIE-Light-DT-1.0 in Muse Code
Copy an example into Muse Code, or ask in your own words.
Use ERNIE-Light-DT-1.0 with maximum reasoning effort to investigate duplicate submissions in this project. Trace the request path, identify a reproducible cause, and propose the smallest fix with a regression test.
Use ERNIE-Light-DT-1.0 to compare the API specification in this repository with its handlers and tests. List mismatches with file references, then propose fixes in order of user impact.
Use ERNIE-Light-DT-1.0 to review this branch's changes alongside the related modules. Report concrete correctness problems with a failing example for each, and keep the review under 800 words.
Use ERNIE-Light-DT-1.0 to summarize the architecture described by this project's source and documentation. Explain the main data flow, identify conflicting documentation, and cite the relevant files.
Use ERNIE-Light-DT-1.0 with maximum reasoning effort to check the retry logic against the project's idempotency requirements. Trace timeout and partial-success cases, then suggest targeted tests for uncovered behavior.
How to set up ERNIE-Light-DT-1.0 in Muse Code
Pick the way that suits you. Each takes about two minutes.
Use Spiral's MCP server
- Install Muse Code
curl -fsSL https://dev.meta.ai/install.sh | bash
- Ask Muse to add the server
Start muse and ask:
Add the Spiral MCP server at https://api.spiral.ai/mcp
- Sign in
In your terminal, run:
muse mcp login spiral
Use the Spiral skill
- Send Muse 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 ERNIE-Light-DT-1.0
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": "baidu/ernie-light-dt-1.0-intl", "input": { "messages": [ { "role": "system" } ] } }'
Then paste one of the prompts above: Muse Code finds ERNIE-Light-DT-1.0 on Spiral, shows the price and runs it.
ERNIE-Light-DT-1.0 providers and prices
Spiral runs ERNIE-Light-DT-1.0 through OpenRouter. Open a card for today's price, with every fee included.
Up to $0.021032 per run at the default inputs. $1.469 per 1M input tokens, $5.122 per 1M output tokens.
Settings you can ask for
Ask Muse Code for any of these in your prompt.
| Setting | Allowed values | Default | Description |
|---|---|---|---|
| Top p | 0 to 1 | ||
| Max tokens | 1 to 131072 | 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 | ||
| Reasoning effort | none, low, high, or max | max | How much the model reasons. |
Every API field of Baidu: ERNIE-Light-DT-1.0 is on its tool page: open its card above.
What can you do with ERNIE-Light-DT-1.0 in Muse Code?
Use ERNIE-Light-DT-1.0 to trace bugs across related code and compare implementations with project requirements. Ask for file references, a focused patch, and evidence from an appropriate test.
Answer questions across long documents
Give ERNIE-Light-DT-1.0 labeled reference text and ask it to answer a specific question with supporting section references and unresolved gaps.
Summarize meeting transcripts
Ask ERNIE-Light-DT-1.0 to turn transcript text into decisions, assigned actions, deadlines, and open questions. Require it to mark missing owners or dates explicitly.
Compare document revisions
Supply the text of two revisions to ERNIE-Light-DT-1.0 and request a table of changed requirements, exceptions, and questions for the author.
Review code with surrounding context
Use ERNIE-Light-DT-1.0 to inspect related source code, trace a suspected defect, and propose a focused fix with a test that reproduces the failure.
Extract records from batches of text
Give ERNIE-Light-DT-1.0 support messages or issue descriptions and request consistent records containing category, evidence, and missing information.
Check plans against constraints
Ask ERNIE-Light-DT-1.0 to compare a proposed schedule or allocation with explicit constraints, identify conflicts, and explain a corrected proposal.
Continue a document review conversation
Use ERNIE-Light-DT-1.0 for follow-up questions about supplied reference text, carrying forward the agreed definitions and asking it to flag contradictions.
More on ERNIE-Light-DT-1.0 and Muse Code
Overview
ERNIE-Light-DT-1.0 in other agents
Questions about ERNIE-Light-DT-1.0 in Muse Code
Can Muse Code use ERNIE-Light-DT-1.0?
Yes. Add Spiral to Muse Code as an MCP server, and Muse Code can find and run ERNIE-Light-DT-1.0 like any built-in tool.
How much does ERNIE-Light-DT-1.0 cost in Muse Code?
Up to $0.021032 per run at the default inputs. $1.469 per 1M input tokens, $5.122 per 1M output tokens. You pay from your Spiral balance as you go, with no subscription, and a failed run is not charged. Your Muse Code plan is separate.
Do I need an OpenRouter account to use ERNIE-Light-DT-1.0 in Muse Code?
No. Spiral runs ERNIE-Light-DT-1.0 on its own OpenRouter account. You sign in to Spiral once and never paste a key.
How long does it take to set up ERNIE-Light-DT-1.0 in Muse Code?
About two minutes: install Muse Code, then ask Muse to add the server, then sign in.
Does Muse Code wait while ERNIE-Light-DT-1.0 runs?
ERNIE-Light-DT-1.0 runs in the background. Muse Code starts the run, checks on it and gives you the text, json when it's ready.
What is ERNIE-Light-DT-1.0 useful for?
Baidu lists long-document questions, summaries, multi-turn conversations, and batch text processing as intended uses. Coding and math or logic problems are also reasonable tasks to try, based on Baidu's stated strengths.
How much context can ERNIE-Light-DT-1.0 handle?
The published context window is 1,024,000 tokens. For document questions, label each source and request section references so you can check the answer against the supplied text.
How long can an answer be?
The published completion limit is 131,072 tokens. Reasoning also consumes output tokens, so that limit does not promise an answer containing 131,072 tokens of visible text.
Can ERNIE-Light-DT-1.0 read images or produce video?
Baidu documents it as text in and text out. Supply extracted text for document analysis; choose a model with image or video support for visual tasks.
When should I use more reasoning effort?
Try maximum effort for complex debugging, constraint checking, or tasks with several dependent decisions. Compare the result with a lighter setting on representative examples, since additional reasoning consumes output tokens.
Start using ERNIE-Light-DT-1.0 in Muse Code
One Spiral account for every tool your agent needs. No subscription; you pay per run.
Paste this into Muse Code:
set up https://spiral.ai/SKILL.md


