Hello! Please tell me. I am creating a chat on my website and I need to connect an existing assistant to this chat via the API. Please help me with the code. I’ve already dug through everything, but the documentation only contains the code that creates a new assistant, but how do I connect to an existing one? Many thanks for the help and maybe an example of the code design!
1 Like
Welcome @analiticfindchia
If you already have an existing assistant you created using Playground or API, you simply have to create a thread, add messages to it, and then run the thread on the assistant.
3 Likes
Could you give an example of a connection. I’m not getting anywhere.
What all have you tried so far? After creating the new assistant, you will get an assistant “Id”
Next, you can use it with a create thread & then create messages in that thread or, do both at the same time.
If you are just starting out, I’d recommend you use the playground to add files (if you need them) & create assistants. It is much faster that way.
Hope this helps
1 Like
Here is my code that I use, and it does not want to work in any way, giving out a 500 error. I don’t know what to do with it anymore. You can tell me where to look. Thank you for your time! I have changed the Api key and the assistant id, they are not valid in this example.
from flask import Flask, request, jsonify, send_from_directory
import requests
import logging
app = Flask(__name__)
API_KEY = ''
ASSISTANT_ID = 'asst_VEYREjiCdv0bTBiqfwuyEywb'
logging.basicConfig(level=logging.DEBUG)
def create_thread():
url = f"https://api.openai.com/v1/assistants/{ASSISTANT_ID}/threads"
headers = {
'Authorization': f'Bearer {API_KEY}',
'Content-Type': 'application/json'
}
data = {
'title': 'New Thread'
}
try:
response = requests.post(url, headers=headers, json=data)
response.raise_for_status()
logging.debug(f"Thread Response JSON: {response.json()}")
return response.json()['id']
except requests.exceptions.RequestException as e:
logging.error(f"Thread creation failed: {e}")
return None
def get_assistant_response(thread_id, user_message):
url = f"https://api.openai.com/v1/assistants/{ASSISTANT_ID}/threads/{thread_id}/messages"
headers = {
'Authorization': f'Bearer {API_KEY}',
'Content-Type': 'application/json'
}
data = {
'messages': [
{'role': 'user', 'content': user_message}
]
}
try:
response = requests.post(url, headers=headers, json=data)
response.raise_for_status()
logging.debug(f"Message Response JSON: {response.json()}")
return response.json()['choices'][0]['message']['content']
except requests.exceptions.RequestException as e:
logging.error(f"Message request failed: {e}")
return "Произошла ошибка при получении ответа от ассистента."
@app.route('/')
def home():
return send_from_directory('', 'index.html')
@app.route('/chat', methods=['POST'])
def chat():
user_message = request.json.get('message')
if not user_message:
return jsonify({"error": "Сообщение пользователя не предоставлено."}), 400
thread_id = create_thread()
if not thread_id:
return jsonify({"error": "Не удалось создать поток."}), 500
assistant_response = get_assistant_response(thread_id, user_message)
return jsonify({"response": assistant_response})
if __name__ == '__main__':
app.run(host='0.0.0.0', port=5000)
@analiticfindchia - Welcome to the community.
For detailed guide by OpenAI click here.
STEP 1 :
Retrieve the assistant id once created using.
assistant = self.client.beta.assistants.create(
name=name,
instructions=instructions,
model="gpt-35-turbo",
temperature= temperature,
top_p= top_p,
tools=tools
)
print(assistant.id) #Prints the assistant id created.
STEP 2:
Create a new thread
thread = client.beta.threads.create()
STEP 3:
Add messages to the thread using this.
message = client.beta.threads.messages.create(
thread_id= thread.id, role="user", content="hi!"
) #change content to the message you receive from the api
STEP 4:
This message on the thread needs to be passed to the LLM for it to generate output. You can use streaming for a typewriter effect or without streaming. Here is a sample code for without streaming.
Note : Handle different run status, I just did two for demonstration. you can find run status desc here
run = self.client.beta.threads.runs.create_and_poll(
thread_id=self.thread_id, assistant_id=self.assistant_id
)
tool_outputs = []
# If the run requires action, process the tools
if run.status == "requires_action":
try:
# Iterate through the tools and execute them
for tool in run.required_action.submit_tool_outputs.tool_calls:
args = json.loads(
run.required_action.submit_tool_outputs.tool_calls[0].function.arguments
)
# If the tool has a name, execute it
if tool.function.name:
print("Invoking tool:", tool.function.name)
function_name = tool.function.name
"""
Call your tools here and save op
"""
# Save the tool output
tool_outputs.append(
{
"tool_call_id": tool.id,
"output": json.dumps(op),
}
)
# If there are tool outputs, submit them
if tool_outputs:
try:
# Submit the tool outputs
run = self.client.beta.threads.runs.submit_tool_outputs_and_poll(
thread_id=self.thread_id,
run_id=run.id,
tool_outputs=tool_outputs,
)
except Exception as e:
print("Failed to submit tool outputs:", e)
else:
print("No tool outputs to submit.")
# If the run is completed, return the messages
if run.status == "completed":
# Get the messages from the thread
messages = self.client.beta.threads.messages.list(
thread_id=self.thread_id
)
return [messages.data[0].content[0].text.value]
STEP 5:
Pass this message back as a response from the api as assistant message.
Cheers 
!
3 Likes
I encountered the same issue, and I believe the method you’re looking for is this:
client = OpenAI()
assistant= client.beta.assistants.retrieve("Your_Assistant_ID")
Thanks @sps for your clear response. It helped me get there.
Additionally, if you are hitting the REST API rather than using the SDK, you need to poll the run, then when it completes, get the messages from the thread filtered by the run_id. The SDK does heavy lifting, but docs are out of sorts right now, so the full hello world boiled down to these steps…
const start = async () => {
// Step 1 - Create a Thread
console.log('Creating thread...');
const threadId = await createThread();
// ...
if (threadId) {
// Step 2 - Add a message to the thread
console.log(`Adding message to thread ${threadId}`);
await addMessageToThread(threadId, MESSAGE);
// Step 3 - Run the assistant on the thread
console.log(`Running thread...`);
let run = await runThreadOnAssistant(threadId);
// Step 4 - Poll the run until finished
console.log(`Polling run ${run.id}...`);
const lastRun = await pollRunUntilFinished(run);
console.log(`Run status: ${lastRun.status}`);
// ...
if (lastRun && lastRun.status === 'completed') {
// Step 5 - Get the response message
console.log(`Fetching assistant response...`);
const assistantResponse = await getResponseMessage(lastRun);
console.log(assistantResponse);
}
}
};
1 Like
Thanks for coming back to confirm!
Hopefully, this thread helps someone else in the future.
1 Like