I am using OpenAI’s assisstant API to make a bot, but the prompt_token_usage in usage is too high.

Run 1 : user: what is your purpose? (it’s in the instructions, instructions are 131 token.)
usage: { prompt_tokens: 7399, completion_tokens: 63, total_tokens: 7462 }
Run 2 : user: what day is it today, based on the internet?
usage: { prompt_tokens: 14631, completion_tokens: 72, total_tokens: 14703 },

async function getSearchResult(query) {
    console.log('------- CALLING AN EXTERNAL API ----------')
    const json = await getJson({
        engine: "google",
        api_key: SERPAPI_KEY,
        q: query,
        location: "Austin, Texas",
    });

    return json["organic_results"];
}

async function checkingStatus(res, threadId, runId, resolve) {
    const runObject = await openai.beta.threads.runs.retrieve(
        threadId,
        runId
    );

    const status = runObject.status;
    console.log(runObject);
    console.log('Current status: ' + status);

    if (status === 'completed') {
        clearInterval(pollingInterval);

        const messagesList = await openai.beta.threads.messages.list(threadId);
        let messages = [];

        messagesList.body.data.forEach(message => {
            messages.push(message.content);
        });

        // Assuming you want the first message in the messages array
        const assistantMessage = messages.length > 0 ? messages[0] : null;

        resolve(assistantMessage);
    }
    // + Addition for function calling
    else if(status === 'requires_action') {
        console.log('requires_action.. looking for a function')

        if(runObject.required_action.type === 'submit_tool_outputs' && status !== 'queued') {
            console.log('submit tool outputs ... ')
            const tool_calls = runObject.required_action.submit_tool_outputs.tool_calls
            const parsedArgs = JSON.parse(tool_calls[0].function.arguments);
            console.log('Query to search for: ' + parsedArgs.query)

            const apiResponse = await getSearchResult(parsedArgs.query)

            const run = await openai.beta.threads.runs.submitToolOutputs(
                threadId,
                runId,
                {
                    tool_outputs: [
                        {
                            tool_call_id: tool_calls[0].id,
                            output: JSON.stringify(apiResponse)
                        },
                    ],
                }
            )
        }
    }
}

app.post('/message', (req, res) => {
    const { message, threadId } = req.body;
    addMessage(threadId, message).then(() => {
        // res.json({ messageId: message.id });

        // Run the assistant
        runAssistant(threadId).then(run => {
            const runId = run.id;

            // Check the status with polling and use a promise to handle completion
            new Promise((resolve) => {
                pollingInterval = setInterval(() => {
                    checkingStatus(res, threadId, runId, resolve);
                }, 5000);
            }).then(async (assistantMessage) => {
                res.json({message: assistantMessage});
            });
        });
    });
});