Skip to content

fix: use original jina de model instead of fp16 due to onnxruntime up… #623

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
joein merged 2 commits into from
Apr 15, 2026
+4 −18
Merged

fix: use original jina de model instead of fp16 due to onnxruntime up… #623

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
2 commits
Select commit Hold shift + click to select a range
0982dbf
fix: use original jina de model instead of fp16 due to onnxruntime up…
joein Apr 8, 2026
85514e6
fix: update size in description
joein Apr 8, 2026
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
4 changes: 2 additions & 2 deletions fastembed/text/pooled_normalized_embedding.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -57,9 +57,9 @@
"Prefixes for queries/documents: not necessary, 2024 year."
),
license="apache-2.0",
size_in_GB=0.32,
size_in_GB=0.64,
sources=ModelSource(hf="jinaai/jina-embeddings-v2-base-de"),
model_file="onnx/model_fp16.onnx",
model_file="onnx/model.onnx",
),
DenseModelDescription(
model="jinaai/jina-embeddings-v2-base-code",
Expand Down
18 changes: 2 additions & 16 deletions fastembed/text/text_embedding.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -90,23 +90,9 @@ def __init__(
**kwargs: Any,
):
super().__init__(model_name, cache_dir, threads, **kwargs)
if model_name.lower() == "nomic-ai/nomic-embed-text-v1.5-Q".lower():
if model_name.lower() == "jinaai/jina-embeddings-v2-base-de":
warnings.warn(
"The model 'nomic-ai/nomic-embed-text-v1.5-Q' has been updated on HuggingFace. Please review "
"the latest documentation on HF and release notes to ensure compatibility with your workflow. ",
UserWarning,
stacklevel=2,
)
if model_name.lower() in {
"sentence-transformers/paraphrase-multilingual-MiniLM-L12-v2".lower(),
"thenlper/gte-large".lower(),
"intfloat/multilingual-e5-large".lower(),
"sentence-transformers/paraphrase-multilingual-mpnet-base-v2".lower(),
}:
warnings.warn(
f"The model {model_name} now uses mean pooling instead of CLS embedding. "
f"In order to preserve the previous behaviour, consider either pinning fastembed version to 0.5.1 or "
"using `add_custom_model` functionality.",
"The model 'jinaai/jina-embeddings-v2-base-de' used to run with fp16 model, but due to onnxruntime updates, now it runs with the original fp32 model.",
UserWarning,
stacklevel=2,
)
Expand Down
Loading