Upload README.md with huggingface_hub

README.md

@@ -111,4 +111,11 @@ Note, v4 is trained with Matryoshka embeddings, and converting to GGUF doesn't b
 
 ### Quantizations
 
-We use [`llama-quantize`](./quantize.sh) with `imatrix` to quantize models from float16. `imatrix` is generated by `llama-imatrix -m jina-embeddings-v4-text-retrieval-F16.gguf -f calibration_data_v5_rc.txt -ngl 99 --no-ppl -o imatrix-retrieval-512.dat`. `calibration_data_v5_rc.txt` can be found [here](https://gist.github.com/tristandruyen/9e207a95c7d75ddf37525d353e00659c/) and is recommended by Unsloth docs.
+We use [`llama-quantize`](./quantize.sh) with `imatrix` to quantize models from float16. `imatrix` is generated by `llama-imatrix -m jina-embeddings-v4-text-retrieval-F16.gguf -f calibration_data_v5_rc.txt -ngl 99 --no-ppl -o imatrix-retrieval-512.dat`. `calibration_data_v5_rc.txt` can be found [here](https://gist.github.com/tristandruyen/9e207a95c7d75ddf37525d353e00659c/) and is recommended by Unsloth docs.
+
+
+Here's the speed and quality evaluation on two nano benchmarks. The higher the better.
+
+![](jina-embeddings-v4-text-retrieval-GGUF on L4.svg)
+![](NanoFiQA2018.svg)
+![](NanoHotpotQA.svg)