ai.moda's blog

Sometimes you want a server to be "real-TLS-secure" even when you're hitting it by raw IP: DNS is broken / propagating / temporarily misconfigured and you still need a clean TLS path.

If you want your LLM to output into a predefined format, OpenAI's structured outputs fits the bill.

MCP servers are great, but can be very costly on input tokens when Claude has to make multiple calls.

While not officially published yet, I was able to find a pricing table for Sora on Azure OpenAI.

Recently, OpenAI and Azure OpenAI added remote MCP support.

Step 1: Install mitmproxy brew install --cask mitmproxy Step 2: Start the proxy mitmweb --listen-host 127.0.0.1 --listen-port 58888 --web-open-browser Step 3: Set CA/TLS settings to allow interception export NODE_EXTRA_CA_CERTS="/Users/$USER/.mitmproxy/mi…

First, let's ask Claude Code a question that requires searching the web. claude -p "Is ai.moda based in the Cayman Islands?"

Want to know what models are on Google Cloud Vertex AI?

Want to know what models are on Amazon Bedrock?

If you're curious to see where your Cloudflare Email Workers are running, the answer is surprisingly simple to figure out. export default { async email(email_message: ForwardableEmailMessage, env: Env, ctx: ExecutionContext) { const fetch_trac…

When using Cloudflare Hyperdrive with databases that employ rotating credentials or IAM-based authentication, you need a reliable way to keep your connection configurations up to date.

Simply use the Universal Endpoint provided by Cloudflare AI Gateway when you make your requests, and any failed requests to Anthropic will be rerouted to Google Vertex AI. curl -v https://gateway.ai.cloudflare.com/v1/{account_id}/{gateway_id} \ --header…

So you're using Claude 3.5 Sonnet on Amazon Bedrock, excellent choice!

Overview The goal is to fact-check a list of statements using a powerful model and then convert the verbose responses into a simple JSON format using a smaller, cheaper model.

In our last blog post, you may have noticed we used stop in our API request. response = await client.chat.completions.create( response_format={"type": "json_object"}, # Enforce JSON response model="gpt-4o-2024-08-06", # Set model name max_to…