Add Requesty preset#752
Conversation
|
Hey, thank you for adding this in. Requesty seems to have a good free tier so a Github Actions verification would be really good thing to have so it won't break so easily. Could you add that in? You can check how the other providers do the actions and copy paste from there |
Adds a tests/e2e_requesty.rs suite mirroring the OpenRouter e2e, plus a test-requesty label-gated GitHub Actions workflow that runs it against the live Requesty endpoint using a REQUESTY_API_KEY secret. Also extends the shared e2e DSL with requesty judge/builder helpers and documents the new label in the README.
|
Added it. There's now a I ran the suite locally against the live Requesty endpoint and it passes (unminifies the fixture and the judge returns EXCELLENT/GOOD). You'll just need to add the |
|
Great, thank you so much! I'll check this out and add the API key to Github Actions on next week |
jehna
left a comment
There was a problem hiding this comment.
💎
I'll verify with Github Actions when I get to my machine from holidays
Adds a
requestypreset, mirroring the existingopenrouterpreset.Requesty (https://requesty.ai) is an OpenAI-compatible LLM router — base URL
https://router.requesty.ai/v1,provider/modelnaming,Authorization: Bearerauth viaREQUESTY_API_KEY.Changes:
src/cli/requesty.rs(new): a faithful copy ofsrc/cli/openrouter.rswithDEFAULTSset to the Requesty base URL,REQUESTY_API_KEY,ProviderKind::OpenAICompat, default modelopenai/gpt-4o-mini(verified live).src/cli/mod.rs: module declaration + re-export.src/main.rs:Requestyclap subcommand variant +into_requesty_args+ dispatch arm.README.md: usage synopsis + a Requesty mode section.Verified with
cargo +1.92 build(Finished, exit 0),cargo fmt --checkclean, and therequesty_defaults_constantsunit test passes. (The repo's deps require rustc ≥1.92; I built with that toolchain.)I work at Requesty. This mirrors the existing OpenRouter provider as closely as possible. Happy to adjust or close it if it's not a fit.