Whenever I make more than one request at a time using the Experts library I get an error stating that my request was aborted.
Error: Request was aborted.
at OpenAI.makeRequest (file:///Users/jamesonparker/Documents/keycollector/key-collector-api/node_modules/openai/core.mjs:289:23)
at process.processTicksAndRejections (node:internal/process/task_queues:95:5)
at async AssistantStream._createAssistantStream (file:///Users/jamesonparker/Documents/keycollector/key-collector-api/node_modules/openai/lib/AssistantStream.mjs:213:24)
at async AssistantStream._runAssistantStream (file:///Users/jamesonparker/Documents/keycollector/key-collector-api/node_modules/openai/lib/AbstractAssistantStreamRunner.mjs:200:16)
This is the code that I'm running.
const database = require('../../database');
let Thread;
let Assistant;
let characterDetailAssistant;
let instructions;
(async () => {
instructions = `.....instructions placeholder.....`
// Initialize Assistant and Thread globally
const Experts = await import('experts');
Thread = Experts.Thread;
Assistant = Experts.Assistant;
myAssistant = await Assistant.create({
id: 'xxxxxxxxxx',
name: 'xxxxxxxxxx',
instructions: instructions,
model: 'gpt-4o-mini',
temperature: 0.1,
response_format: {
"type": "json_object"
}
});
})();
module.exports.parseEbayTitle = async (title) => {
let thread = await Thread.create();
let response = await myAssistant.ask(title, thread.id);
response = JSON.parse(response);
return response;
}
This is used in a job server where I queue up multiple jobs. If I have the job server queue set to run anything more than 1 job at a time I start to get the "Request Aborted" errors. My rate limit is at 5,000 RPM and I'm getting about 1 request back a second so I'm not even close to the lim it here.
Any ideas?
Whenever I make more than one request at a time using the Experts library I get an error stating that my request was aborted.
Error: Request was aborted.
at OpenAI.makeRequest (file:///Users/jamesonparker/Documents/keycollector/key-collector-api/node_modules/openai/core.mjs:289:23)
at process.processTicksAndRejections (node:internal/process/task_queues:95:5)
at async AssistantStream._createAssistantStream (file:///Users/jamesonparker/Documents/keycollector/key-collector-api/node_modules/openai/lib/AssistantStream.mjs:213:24)
at async AssistantStream._runAssistantStream (file:///Users/jamesonparker/Documents/keycollector/key-collector-api/node_modules/openai/lib/AbstractAssistantStreamRunner.mjs:200:16)
This is the code that I'm running.
This is used in a job server where I queue up multiple jobs. If I have the job server queue set to run anything more than 1 job at a time I start to get the "Request Aborted" errors. My rate limit is at 5,000 RPM and I'm getting about 1 request back a second so I'm not even close to the lim it here.
Any ideas?