A powerful batch processing system for executing large-scale AI inference requests asynchronously with significant cost savings.
Install & Download
1. ClawHub CLI
The fastest way to install a skill directly from the registry.
npx clawhub@latest install doubleword
2. Manual Installation
Copy the skill folder to one of these locations
~/.openclaw/skills/ <project>/skills/ Priority: Workspace > Local > Bundled
3. Prompt Installation
Copy this prompt to OpenClaw to install it automatically.
Help me install doubleword using Clawhub. If Clawhub is not installed, install it first (npm i -g clawhub).
Prefer to download?
Get the raw skill files in a ZIP archive.
What is Doubleword Batches?
Doubleword Batches enable developers to manage high-volume AI workloads through an asynchronous API designed for maximum throughput and efficiency. By utilizing this skill within the Openclaw Skills ecosystem, users can submit thousands of requests in a single JSONL file, allowing the system to process them in the background. This approach is particularly valuable for non-real-time tasks where cost efficiency is a priority, offering savings of up to 83% compared to standard real-time inference.
The skill supports advanced flagship models like Qwen3-VL and maintains full compatibility with OpenAI batch protocols. This ensures that developers can easily integrate tool calling and structured outputs into their batch workflows without changing their existing logic or data structures.
Doubleword Batches Use Cases
- Processing massive datasets for sentiment analysis or categorization where immediate responses aren't required.
- Executing large-scale content generation tasks while remaining 50-80% under standard API budgets.
- Scaling AI agents that need to perform thousands of tool-calling operations asynchronously.
- Migrating high-volume OpenAI batch workloads to a more cost-effective infrastructure.
- Extracting structured data from thousands of documents using specific JSON Schemas.
How Doubleword Batches Works
- Generate a JSONL file where each line represents a single AI completion request with a unique custom ID.
- Upload the request file to the Doubleword storage system to obtain a reference file ID.
- Create a batch job by specifying the input file ID and choosing a completion window (1-hour or 24-hour SLA).
- Poll the batch status endpoint to monitor progress as the system validates and processes the requests.
- Download the completed results file which maps every response back to the original custom ID for easy data processing.
Doubleword Batches Setup
To use this skill with Openclaw Skills, you must first create an account at app.doubleword.ai and generate an API key.
Install CLI Helpers
If you wish to use the automatic batching client:
pip install autobatcher
Submit a Batch via Bash
curl https://api.doubleword.ai/v1/files \
-H "Authorization: Bearer $DOUBLEWORD_API_KEY" \
-F purpose="batch" \
-F file="@my_requests.jsonl"
Doubleword Batches Data Schema & Taxonomy
The skill processes data using the JSON Lines (JSONL) format. Each request line must be a valid JSON object.
| Field | Requirement | Description |
|---|---|---|
| custom_id | Required | A unique string (max 64 chars) to identify the request in the output. |
| method | Required | The HTTP method to use (must be POST). |
| url | Required | The API endpoint, typically /v1/chat/completions. |
| body | Required | The standard payload including model, messages, and parameters. |
Upon completion, the output file provides a response object for every successfully processed line, while errors are logged in a separate error file for easy debugging.
Doubleword Batches Advanced Features
- Automatic batching via the autobatcher Python client for transparently converting individual calls into batches.
- Support for complex Tool Calling, allowing models to interact with external function definitions at scale.
- Guaranteed Structured Outputs via JSON Schema to ensure model responses follow strict data formats.
- Cost-optimized SLAs including 1-hour and 24-hour windows for massive pricing discounts.
- Early access retrieval, enabling users to download partial results via the X-Incomplete header before the full batch finishes.
SKILL.md
Loading
Related Openclaw Skills
Doubleword Batches
A high-performance batch processing skill for running asynchronous AI inference jobs at scale through the Doubleword API.
mflux: Local Apple Silicon Image Generation
A high-performance MLX-based skill for generating images locally on Apple Silicon using FLUX.2 and Z-Image models.
ModelWar AI CoreWar Arena
A competitive arena where AI agents battle in a virtual computer using the classic CoreWar programming game.
Scrapling
A high-performance Python-based web scraping tool designed for resilient data extraction from static, dynamic, and anti-bot protected websites.
Baselight MCP
A powerful MCP server connection providing real-time SQL access to over 50 premium datasets directly within your AI coding agent or IDE.
Cochesnet CLI
A command-line tool designed to search the coches.net marketplace and retrieve comprehensive listing details in table or JSON formats.
Featured*
