Aaron Pham

Results 107 issues of Aaron Pham

This is probably not a bug, but I'm just wondering whether there is a decision not to make docstring for metrics RST-friendly. Let's take the documentation for https://github.com/prometheus/client_python/blob/db391dbc83df4202698a21463ae89d6fb3611ce2/prometheus_client/metrics.py#L516 as an...

🎉 OpenLLM 🤝 LangChain [OpenLLM](https://github.com/bentoml/OpenLLM) is a new open platform for operating large language models(LLMs) in production. Serve, deploy, and monitor any LLMs with ease. OpenLLM lets developers and researchers...

### Feature request openllm.LLMConfig should be serializable as a dict ```python config = openllm.AutoConfig.for_model('dolly-v2') dict(config) # should be the same as config.model_dump(flatten=True) ``` This means within the `LLMConfig` class generation,...

### Describe the bug This is rather a feature, where parsing configuration should support union type. TODO: - convert union to correct click type. This probably will involve writing custom...

### Feature request This ticket keeps track of implementation support for each of the adapter type for models under OpenLLM: - OPT: Lora - Dolly-v2: wip - Flan-T5: wip -...

performance

A custom strategies that brings support for AMD GPU (experimental) Address #65 Signed-off-by: Aaron

Signed-off-by: Aaron