[HN Gopher] AIConfig Extension for VS Code
___________________________________________________________________
AIConfig Extension for VS Code
Author : cybereporter
Score : 85 points
Date : 2024-03-08 17:00 UTC (6 hours ago)
(HTM) web link (marketplace.visualstudio.com)
(TXT) w3m dump (marketplace.visualstudio.com)
| bongodongobob wrote:
| I don't understand what the use case is here at all. It saves
| prompt parameters as JSON? Is that it?
| cantSpellSober wrote:
| Using models from any provider or modality in a single
| playground
| cybereporter wrote:
| JSON and YAML, but it also supports connectivity to a bunch of
| LLMs (including image models like Dall-e 3) and some prompt
| chaining.
| m3kw9 wrote:
| Looks like you can switch models and use the same prompts very
| quickly.
| slig wrote:
| If you enjoy the OpenAI's playground, this looks pretty much
| like that but you can use other LLMs.
| skybrian wrote:
| It seems to be a config file format, library for loading it,
| interactive editor for it, and playground that uses it.
| leguy wrote:
| I thought it supported a few use cases:
|
| * Testing prompt behavior across various LLMs * Sharing those
| prompts across multiple applications
|
| We currently use a jupyter notebook to iterate, test, and
| validate prompts. Then move those prompts to our production app
| written in C#. If there were a C# SDK, I could use this tool to
| create a prompts config file and share it between the jupyter
| notebook and the C# app. The config file could also be added to
| version control.
|
| Having said that, I don't understand why it saves the output of
| the LLM so maybe I'm missing something.
|
| https://aiconfig.lastmileai.dev/docs/basics
| huac wrote:
| VSCode is a good target for single-player editing, and I can see
| something like this being helpful, but what does a collaborative
| experience look like, eg if you have a team of folks all working
| on the same prompts?
| cybereporter wrote:
| Yeah, source controlling the configurations, then cross-
| collaboration on teams isn't supported, but something
| interesting to explore especially with engineers and PMs
| collaborating together on prompting.
| TommyDANGerous wrote:
| So one challenge we've had is folks would copy paste their
| prompt in google docs then write under it then try to run it
| and then paste back what worked well. A collaborative
| experience would be super useful so we don't have to jump out
| and can just append/build on each other's prompts inline.
| dumbo-octopus wrote:
| Isn't that what Git is for? Or do you mean lot's of people
| editing the same prompt at the same moment, a la google docs?
| That sounds... hectic.
| awildjia wrote:
| So I guess the idea is that you do your experimenting and
| iteration with this, then plug the config into your app?
| cybereporter wrote:
| Yeah, it seems like it. It's nice to have a way to share your
| prompts + logic without asking others to compile and build your
| app.
| sa-code wrote:
| It contains telemetry but it's not clear whether it collects your
| prompts or not
| cybereporter wrote:
| doesn't collect prompts and there's a way to disable telemetry
| as well - https://github.com/lastmile-
| ai/aiconfig/blob/8a5a59d47cef474...
| dumbo-octopus wrote:
| Alternative extension to do something similar, but using VSCode's
| native notebooks:
| https://marketplace.visualstudio.com/items?itemName=jaaxxx.l....
| It's more tailored to crafting long-form few-shot prompts, rather
| than a single text box. Think ChatGPT playground but with a text
| editor interface that isn't absolutely terrible.
|
| Nice part about that one is no separate server, no telemetry, and
| the backing file format is simple JSON you can directly import
| from your production application. However the range of supported
| models is smaller (basically only LLaMa-style and OpenAI-style
| interfaces are supported).
___________________________________________________________________
(page generated 2024-03-08 23:01 UTC)