[Feature]: Consistent format for model_prices_and_context_window file

[krrishdholakia]

The Feature

Each chat, completion model should have:

• max_input_tokens
• max_output_tokens
• input_cost_per_token or input_cost_per_second
• output_cost_per_token or output_cost_per_second
• llm provider (azure, bedrock, etc.)
• mode (embedding, completion, image generation)

Each embedding model should have:

• max input tokens
• dimensions (e.g. 2048 for text-embeddings)
• input_cost_per_token or input_cost_per_second
• output_cost_per_token or output_cost_per_second
• llm provider (azure, bedrock, etc.)
• mode (embedding, completion, image generation)

Motivation, pitch

If it's consistent, then people can write code on top of it

Twitter / LinkedIn details

No response

[krrishdholakia]

interesting - https://openrouter.ai/api/v1/models