bge-reranker-base GGUF
GGUF format of BAAI/bge-reranker-base for use with CrispEmbed.
BGE Reranker Base. Cross-encoder reranker for English and Chinese. Use with crispembed_rerank().
Files
| File | Quantization | Size |
|---|---|---|
| bge-reranker-base-q4_k.gguf | Q4_K | 247 MB |
| bge-reranker-base-q8_0.gguf | Q8_0 | 287 MB |
| bge-reranker-base.gguf | F32 | 1066 MB |
Quick Start
# Download
huggingface-cli download cstr/bge-reranker-base-GGUF bge-reranker-base-q4_k.gguf --local-dir .
# Run with CrispEmbed
./crispembed -m bge-reranker-base-q4_k.gguf "Hello world"
# Or with auto-download
./crispembed -m bge-reranker-base "Hello world"
Model Details
| Property | Value |
|---|---|
| Architecture | BERT |
| Parameters | 278M |
| Embedding Dimension | 768 |
| Layers | 12 |
| Pooling | CLS |
| Tokenizer | WordPiece |
| Base Model | BAAI/bge-reranker-base |
Verification
Verified bit-identical to HuggingFace sentence-transformers (cosine similarity >= 0.999 on test texts).
Usage with CrispEmbed
CrispEmbed is a lightweight C/C++ text embedding inference engine using ggml. No Python runtime, no ONNX. Supports BERT, XLM-R, Qwen3, and Gemma3 architectures.
# Build CrispEmbed
git clone https://github.com/CrispStrobe/CrispEmbed
cd CrispEmbed
cmake -S . -B build && cmake --build build -j
# Encode
./build/crispembed -m bge-reranker-base-q4_k.gguf "query text"
# Server mode
./build/crispembed-server -m bge-reranker-base-q4_k.gguf --port 8080
curl -X POST http://localhost:8080/v1/embeddings \
-d '{"input": ["Hello world"], "model": "bge-reranker-base"}'
Credits
- Original model: BAAI/bge-reranker-base
- Inference engine: CrispEmbed (ggml-based)
- Conversion:
convert-bert-embed-to-gguf.py
Provenance and EU AI Act Art. 53 note
- Upstream model: BAAI/bge-reranker-base — published by
BAAI. - Upstream licence:
mit. 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 (GGUF/GGML). 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.