Joseph Roddy

Hey @WorkingOnThis Sorry I'm just seeing this! Thanks for opening an issue. Were you able to fix this at all? I'm guessing a new version of Prisma might've broken this...

Thanks for the heads up @Gherdty ! Will take a look when I get a chance, hoping to play around with this project this weekend to see if it works...

Seeing the same issue on my end. Temporarily, if anyone else is running into this, you can cast the type as a workaround: ```ts import { PrismaClientPostgres } from '@prismany/client';...

Should be fixed with version `1.0.2`. Let me know if you get a chance to try it out @Gherdty Tested with Prisma 5 and that seems to be working for...

Thanks for bringing this up @safejace ! Hmm, I don't usually use pnpm, and I've never used pnpx. Can you try adding the the package to your dev dependencies? npx...

I guess it's possible the second error could be a totally new, unrelated bug with prismany. Do you not get those missing files with [email protected]?

Hm, yeah it sounds like an issue with how it's getting bundled with Nuxt specifically. If you just do `pnpx prisma generate`, are you able to import the regular `PrismaClient`?...

Relevant discord discussion from a similar issue. https://discord.com/channels/1035538056146595961/1338578707086049372 I was never able to create a simple proof of concept replication. It was only happening in our large real world project...

I'm able to fix this by setting `SMART_LLM=openai:o4-mini` in my .env, but am wondering if I'm potentially doing something else wrong.

> ..."tokens per min (TPM): Limit 30000, Requested 49372." > > You are overwhelming the model, hitting the limit, OpenAI refuses to answer, the research must fail. > You can...