Darren Shepherd

Yeah, ? only works on built in tools, that wasn't intentional but more of a bug of how I hacked in support for that. We should support it for all...

You can set `OPENAI_URL_BASE` to point to ollama with gptscript today. Pretty much everything is standarizing to the OpenAI chat completion API and ollama supports that. The unfortunate thing is...

Python and Node.js as first priorities.

Try `gptscript --help` ``` --default-model string Default LLM model to use ($GPTSCRIPT_DEFAULT_MODEL) (default "gpt-4-turbo-preview") ``` Also do `gptscript --list-models` to see which models you have available to you. I'm surprised...

Based on your previous issue (and the fact I was just stuck in an airport for 3 hours). I rewrote this script so that it 1) uses less memory 2)...

I just pushed the rewritten script at https://github.com/ibuildthecloud/coreos-on-do/tree/automation I haven't fully tested it yet and will over the weekend. I'll test all the regions too. The main motivation for why...

@aaronogle @mishak87 I just merged the automation branch to master. If you have time it would be nice if you could test it out. I personally tested 512mb droplets in...

The syntax for setting bools to false is `--newApp=false`. What is happening here is that it's being interpreted as one flag `--newApp`, which means set to true, and then an...

This is because CentOS uses device mapper. That graph driver isn't very friendly to Docker in Docker. What happens in that on start it creates 2 loop back devices. When...

Just wanted to throw out there that CentOS 7.2 when released will support overlayfs and then you won't see these issues.