tstadel

feat: hf inference support for gated repos

1

### Related Issues - this allows us to use `meta-llama/Meta-Llama-3-8B-Instruct` via HF inference API ### Proposed Changes: - pass hf token to `DefaultPromptHandler` for tokenizer ### How did you test...

feat: extend PromptBuilder and deprecate DynamicPromptBuilder

21

### Related Issues Currently we cannot have both: - a default prompt template defined (PromptBuilder) - dynamically change prompt templates at runtime (DynamicPromptBuilder) There are two options: - **A** we...

feat: add ChatPromptBuilder, deprecate DynamicChatPromptBuilder

1

### Related Issues - follow up from https://github.com/deepset-ai/haystack/pull/7655 regarding `ChatPromptBuilder` ### Proposed Changes: This extends ChatPromptBuilder to change prompts at query time. ```python default_template = [ChatMessage.from_user("This is the default prompt:...

Currently some templates have version 0.0.1, some 0.1.0 and one has 0.1.1 . It's unclear what that even means. We should unify the versions (e.g. all 1.0.0) if there's no...

### Related Issues - fixes raising the following error if running `prompt_node.run(query="", documents=documents)` ``` Exception while running node 'PromptNode': Expected prompt parameters ['documents', 'query'] to be provided but got only...

fix: meta from ByteStream input for AzureOCRDocumentConverter

1

### Related Issues - fixes #issue-number ### Proposed Changes: ### How did you test it? ### Notes for the reviewer ### Checklist - I have read the [contributors guidelines](https://github.com/deepset-ai/haystack/blob/main/CONTRIBUTING.md) and...

### Related Issues - support for Triton-hosted embedding models: e.g. almost any hugginface model can be hosted on [Triton](https://developer.nvidia.com/triton-inference-server) and thus make use of Triton's hosting and scaling features. This...

feat: extend component run info

7

### Related Issues - fixes logging component run info via logging extra in addition to tracing + adds additional information (e.g. input and output lengths). Background: Currently there is little...

feat: streaming_callback as run param from HF generators

1

### Related Issues - `streaming_callback` run param should be supported in a unified way throughout haystack main package. So far only `OpenAIGenerator` and `AzureOpenAIGenerator` support the run param. `HuggingFaceAPIGenerator` and...