trithemius/granite-embedding-reranker-english-r2-GGUF

granite-embedding-reranker-english-r2-GGUF

GGUF F16 conversion of `ibm-granite/granite-embedding-reranker-english-r2`.

Model Summary

This repository provides an F16 GGUF export of the original IBM Granite English reranker / embedding model for use in GGUF-compatible tooling where supported.

• —Base model: ibm-granite/granite-embedding-reranker-english-r2
• —Format: GGUF
• —Precision: F16
• —Quantization: None
• —Source: Hugging Face original model repo by IBM Granite

This repository is a format conversion of the original model. Please refer to the original model card for authoritative details on training data, intended uses, limitations, and evaluation.

Files

• —granite-embedding-reranker-english-r2-f16.gguf — GGUF model in full F16 precision

Provenance

• —Original model: `ibm-granite/granite-embedding-reranker-english-r2`
• —Converted by: YOUR_USERNAME
• —GGUF conversion tool: llama.cpp / convert_hf_to_gguf.py
• —Conversion type: HF → GGUF
• —Output precision: F16

Intended Use

This model is intended for embedding / reranking tasks, subject to support in the target runtime.

Typical use cases may include:

• —semantic ranking
• —retrieval re-ranking
• —passage-query scoring
• —information retrieval pipelines
• —search relevance improvement

Important Compatibility Note

This is not a generative chat model. It is a converted reranker / embedding-style model.

GGUF is most commonly used with runtimes centered around LLM inference. Support for encoder-style, embedding, or reranker architectures may vary depending on the runtime and version.

Before using this file, verify that your target GGUF runtime supports:

1. 1.the underlying model architecture
2. 2.the required pooling / scoring behavior
3. 3.encoder or reranker inference modes if applicable

Example Download

Using huggingface-cli:

huggingface-cli download YOUR_USERNAME/granite-embedding-reranker-english-r2-GGUF \
  granite-embedding-reranker-english-r2-f16.gguf \
  --local-dir .