Alex Reibman

> This looks like a very cool idea, can I work on it? Sure, feel free to take a stab. Are you using DSPY on the regular? Would love to...

> I see this on Langwatch: https://docs.langwatch.ai/dspy-visualization/quickstart Could probably fork the integration approach they used

> Hey, does the new LlamaIndex_callback_handler should mimic the langchain_callback_handler which is already present ? You can read more about the LlamaIndex callback handler here: https://docs.llamaindex.ai/en/stable/module_guides/observability/callbacks/root.html#modules The idea is this:...

Looking at similar event queue-based event trackers patterns (Kafka, Amplitude, Segment, etc.), it looks like there will always be a delay unless the consumer changes their entire event loop to...

Wrote a fix in #21. Basically, introduced retries with exponential backoff. Solves the problem nicely

Potentially, yes. I believe Claude 1 uses cl100k, but Claude 2 and Claude 3 I'm not so sure. Here is the only info I've found on Claude 3: https://github.com/javirandor/anthropic-tokenizer And...

> https://www.lepton.ai/playground > > bunch of models provided : llama7b,13b,70b, codellama 7b,13b,34b, mixtral-8*7b, etc... > > For pricing: https://www.lepton.ai/pricing I'm assuming these are the prices: Model | Price -- |...

> and they are open AI compatible, eg. for mixtral-8*7b https://www.lepton.ai/playground/chat?model=mixtral-8x7b Sounds good. I can add these, but if you'd prefer, raise a PR and I'll approve.

> I'm assuming the structure of [this file](https://github.com/AgentOps-AI/tokencost/blob/main/tokencost/constants.py) needs to be altered a bit. maybe add a provider attribute to each model? and default provider to openAI? or have a...

@bobmayuze I've created a new cost dictionary file: https://github.com/AgentOps-AI/tokencost/blob/main/model_prices.yaml Feel free to add your models in there. Additionally, you can update the table in the README to include those models.