cstr
/

ettin-reranker-1b-v1-GGUF

Model card Files Files and versions

Ettin Reranker 1B — GGUF

GGUF quantizations of cross-encoder/ettin-reranker-1b-v1, a ModernBERT-based cross-encoder reranker (1.00B params, Apache-2.0).

Converted with CrispEmbed.

Files

File	Quant	Size
	F32	4126 MB
	Q8_0	1107 MB
	Q4_K	638 MB

Architecture

Base: ModernBERT (28 layers, 1792 hidden)
Classifier head: Dense(1792→1792, GELU) → LayerNorm → Dense(1792→1)
Tokenizer: GPT-2 ByteLevel BPE (50368 tokens)

Usage

Downloads last month: 147

GGUF

Model size

1B params

Architecture

bert

Hardware compatibility

Log In to add your hardware

6-bit

8-bit

32-bit

Inference Providers NEW

This model isn't deployed by any Inference Provider. 🙋 Ask for provider support

Model tree for cstr/ettin-reranker-1b-v1-GGUF

Base model

jhu-clsp/ettin-encoder-1b

Finetuned

cross-encoder/ettin-reranker-1b-v1

Quantized

(2)

this model