You can deploy this model and start earning money today!

software-mansion/
react-native-executorch-lfm2.5-embedding-350m

Name: react-native-executorch-lfm2.5-embedding-350m
Availability: InStock

executorch

other

Views

No views yet

lfm2.5-embedding-350m

This repository hosts the lfm2.5-embedding-350m models exported for the React Native ExecuTorch library as ExecuTorch .pte programs, ready to run on device.

Upstream model: LFM2.5-Embedding-350M

Variants

Path	Backend	Precision
`mlx/lfm_2_5_embedding_350m_mlx_int4.pte`	mlx	int4
`xnnpack/lfm_2_5_embedding_350m_xnnpack_8da4w.pte`	xnnpack	8da4w

Unshipped backends

A backend directory carrying a NOTES.md and no .pte was exported and not published. Its note records what went wrong.

coreml (why)

Repository structure

config.json                                       43 B
coreml/NOTES.md                                   1.6 kB
mlx/config.json                                   1015 B
mlx/lfm_2_5_embedding_350m_mlx_int4.pte           180 MB
tokenizer.json                                    4.5 MB
tokenizer_config.json                             89.9 kB
xnnpack/config.json                               1.0 kB
xnnpack/lfm_2_5_embedding_350m_xnnpack_8da4w.pte  548 MB

Compatibility

These files are published for the ExecuTorch v1.4.1 runtime. ExecuTorch gives no forward compatibility guarantee, so an older runtime may fail to load them.

To use them in React Native ExecuTorch, pass the model constant shipped in the library's model registry to the corresponding task pipeline. See the documentation.

To load these files in your own ExecuTorch runtime, read the compatibility note first.

Model details

Architecture: LFM2.5-350M bidirectional backbone (hybrid conv + attention, hidden size 1024) + CLS pooling + L2 normalize. The exported graph bakes in CLS pooling and L2 normalization, so the runner consumes (input_ids, attention_mask) and receives the final unit-norm embedding directly.
Output dimension: 1024.
Similarity metric: cosine (embeddings are L2-normalized, so a dot product equals cosine).
Prompts: the model is trained with asymmetric query: / document: text prefixes. Prepend query: to search queries and document: to indexed passages for best retrieval quality.

software-mansion/react-native-executorch-lfm2.5-embedding-350m