Does anyone with 100 AI agents running simultaneously and $10k cursor bill actually ship?

At that point does it make sense to just invest in a local AI setup?

Good question

Do open source models compete with sota right now?

Applied AI, databases, data, vibe coding (sic.), open source, community/devrel, remote. 

itinerant businessperson. globalist.

Nope. Llama is let’s just say not too impressive. Qwen is underrated. I haven’t tried the new Deepseek and I can’t comment on what mistral has given recently either. 

It’s pretty hard to beat Claude, Gemini, and ChatGPT frankly. This is “correct” as of today - this can change later of course.

Even if it were fast enough is it any good compared to the frontier rn?

You can use Deepseek, Qwen, llama, via ollama locally. Or buy a beefy Mac mini/studio. I haven’t found it to be fast enough with what I have but it’s good to use OpenAI compatible things

I haven't been following but it probably depends on the task. I'd be curious to see someone compare local vector db + RAG / mcp for specific tasks compared to big foundation models via API to see if you can build around your data set or problem space for cheaper* or try and figure out how long you'd break even for the 80/20 of your actual use case. Maybe use the expensive stuff as a fallback even

Seems like DeepSeek R1 2805 is up there with the others and you could have it local with some effort