Repository navigation
Added Vercel AI Gateway as a provider - #11
Conversation
|
@jerilynzheng Thanks! Can you just check one thing for me -- someone noticed earlier that with Prime Intellect's inference on the OAI API, there was some mismatch for cost tracking. Can you just check that with Vercel (I don't have an API key so I can't test this) that cost tracking works (e.g. when you run the RLM example, can you print the RLMChatCompletion object and send a screenshot here? Just make sure that the costs it reports are correct). |
|
Hey @alexzhang13 are you thinking of using dspy for managing llm inference given it already has integration with the different provider through litellm |
Haven't though of this yet, could be convinced to add it though. |
|
HI @alexzhang13 , sorry for the delay! Here's the screenshot |
…e review Loader hardening: non-finite S6 values rejected at the caps gate (#1); rejections keyed by unique directory names (#2); declared-id mismatch caught at the text level before any candidate code runs (#3); gate subprocess launch/decode failures return structured rejections (#11); unhashable surface values rejected instead of raising (#12). Ledger contract: non-finite band bounds render as null and all persisted JSON is strict RFC-8259 (#4); rejection-only rounds persist an auditable ledger with a null baseline (#5). Live smoke round regenerated with the fixed serialization. Claude-Session: https://claude.ai/code/session_01QfVmUyopDjKxK3b4nWxKWM
3 participants
Summary
Added Vercel AI Gateway as a provider using OpenAI-compatible API
Changes