File	Size	Description
`model.fp16.onnx`	~1.3 MB	ONNX header (external data) — Constant_2836 inlined 2026-05-03
`model.fp16.onnx.data`	~1.2 GB	FP16 weights
`tokenizer.json`, `config.json`, `tokenizer_config.json`	small	tokenizer + model config

Conversion

Streaming FP32→FP16 via convert_fp16_streaming.py (bypasses the 2 GB protobuf serialization limit).

2026-05-03 fix

The original FP16 export referenced two external data files: model.fp16.onnx.data (uploaded here) and model.onnx_data (which was never uploaded — it lives in the upstream FP32 repo). Only one tensor (Constant_2836, a 256-byte 1×64×1 float — likely a rotary embedding constant) was affected. That tensor's bytes were copied from jinaai/jina-embeddings-v5-text-small-retrieval/onnx/model.onnx_data (offset 2,384,199,680, length 256) and inlined directly into the model.fp16.onnx header. The 1.2 GB .data payload is unchanged.

Use via fastembed-rs

rust

let embedder = TextEmbedding::try_new(
    InitOptions::new(EmbeddingModel::JinaEmbeddingsV5SmallFp16))?;

Pooling: last-token. Asymmetric retrieval prefixes "Query: " / "Document: " are recommended.

License

Apache 2.0, inherited from the base model.

Provenance and EU AI Act Art. 53 note

Upstream model: jinaai/jina-embeddings-v5-text-small-retrieval — published by jinaai.
Upstream licence: cc-by-nc-4.0. This repository redistributes under the same terms; it grants no rights the upstream licence does not.
What was done here: format conversion and/or quantisation only (ONNX, F16 precision). No training, no fine-tuning, no merging, no distillation, no change to architecture, vocabulary or capability. Only the numeric representation of the upstream weights differs.
Training data: documented — where it is documented at all — by the upstream provider; see the upstream model card. No training data was used, added or selected by this repository.
Provider status: under Regulation (EU) 2024/1689 the upstream authors remain the provider of this model. Converting the serialisation format does not make this repository the provider of a new general-purpose AI model, and no such claim is made. Questions about training content, copyright policy or model capability belong upstream.

cstr/jina-embeddings-v5-text-small-retrieval-onnx-fp16

jina-embeddings-v5-text-small-retrieval — FP16 ONNX

Quality

Files

Conversion

2026-05-03 fix

Use via fastembed-rs

License

Provenance and EU AI Act Art. 53 note

cstr/
jina-embeddings-v5-text-small-retrieval-onnx-fp16