phoenix-oss/llama-stack-mirror
0
0
Fork 1
mirror of https://github.com/meta-llama/llama-stack.git synced 2025-07-01 03:58:40 +00:00
Code Issues Projects Releases Packages Wiki Activity Actions

feat: Add synthetic-data-kit for file_search doc conversion

Browse source 
This adds a `builtin::document_conversion` tool for converting
documents when used with file_search that uses
meta-llama/synthetic-data-kit. I also have another local
implementation that uses Docling, but need to debug some segfault
issues I'm hitting locally with that so pushing this first as a
simpler reference implementation.

Long-term I think we'll want a remote implemention here as well - like
perhaps docling-serve or unstructured.io - but need to look more into
that.

This passes the existing
`tests/verifications/openai_api/test_responses.py` but doesn't yet add
any new tests for file types besides text and pdf.

Signed-off-by: Ben Browning <bbrownin@redhat.com>
This commit is contained in:
Ben Browning 2025-06-20 18:09:14 -04:00
parent 6fde601765
commit e56690abef
18 changed files with 230 additions and 18 deletions
Download patch file Download diff file Expand all files Collapse all files

8
llama_stack/providers/registry/vector_io.py
View file

@ -24,7 +24,7 @@ def available_providers() -> list[ProviderSpec]:
config_class="llama_stack.providers.inline.vector_io.faiss.FaissVectorIOConfig",
deprecation_warning="Please use the `inline::faiss` provider instead.",
api_dependencies=[Api.inference],
optional_api_dependencies=[Api.files],
optional_api_dependencies=[Api.files, Api.tool_runtime],
),
InlineProviderSpec(
api=Api.vector_io,
@ -33,7 +33,7 @@ def available_providers() -> list[ProviderSpec]:
module="llama_stack.providers.inline.vector_io.faiss",
config_class="llama_stack.providers.inline.vector_io.faiss.FaissVectorIOConfig",
api_dependencies=[Api.inference],
optional_api_dependencies=[Api.files],
optional_api_dependencies=[Api.files, Api.tool_runtime],
),
# NOTE: sqlite-vec cannot be bundled into the container image because it does not have a
# source distribution and the wheels are not available for all platforms.
@ -44,7 +44,7 @@ def available_providers() -> list[ProviderSpec]:
module="llama_stack.providers.inline.vector_io.sqlite_vec",
config_class="llama_stack.providers.inline.vector_io.sqlite_vec.SQLiteVectorIOConfig",
api_dependencies=[Api.inference],
optional_api_dependencies=[Api.files],
optional_api_dependencies=[Api.files, Api.tool_runtime],
),
InlineProviderSpec(
api=Api.vector_io,
@ -54,7 +54,7 @@ def available_providers() -> list[ProviderSpec]:
config_class="llama_stack.providers.inline.vector_io.sqlite_vec.SQLiteVectorIOConfig",
deprecation_warning="Please use the `inline::sqlite-vec` provider (notice the hyphen instead of underscore) instead.",
api_dependencies=[Api.inference],
optional_api_dependencies=[Api.files],
optional_api_dependencies=[Api.files, Api.tool_runtime],
),
remote_provider_spec(
Api.vector_io,