LLM Eval and Compare

Idk I'll write this later just look at the 2 pages and read the code Idk I'll write this later just look at the 3 pages and read the code

Compare models

Compare prompts

Compare chats

Transcribe audio

TODO v0.1.0

• model/provider selection dropdown

• remove selected model button

• initial results table skeleton

• info button top left on eval page

• info button top left on compare page

• info button top left on transcription page

• header component with the nav dropdown and github icon link

• ? or just add the github link to the dropdown?

• parallel requests and streaming for compare page

• results table outputs should be equal width per model

• results table resizable

• Github logo link to repo

• fix margin of header icons

• add all model options to the dual llm comparison chat

• about page

• feedback / submit bug / feature request

• stt page with Groq whisper

• transcription optional arguments

• lambda chat api support

• hermes 3 models added

• upload 'vibe check list' button

• Markdown rendering for responses

v0.2.0

• auth / sign up / login pages

• save prompt collection

• saved 'vibe check list'

• saved eval runs

v0.3.0

• API access
• batch audio processing

v1.0.0

• create documentation