Upload README.md with huggingface_hub

README.md

@@ -7,7 +7,7 @@ tags:
   - quantized
   - embedding
   - nvidia-modelopt
-  - nanojet
+
 pipeline_tag: feature-extraction
 ---
 
@@ -78,21 +78,6 @@ All configurations achieve >0.99 cosine similarity with the BF16 baseline.
 └── generation_config.json   # Generation config
 ```
 
-## Usage
-
-This model is designed to be used with the [NanoJet](https://github.com/ai-microsoft/NanoJet_Kernels) inference engine:
-
-```python
-from test.infrastructure.model_utils import load_nanojet_model
-
-model = load_nanojet_model(
-    "1kxia/gemma-3-270m-modelopt-fp8",
-    batch=8,
-    seq_len=4096,
-    quantization="fp8"
-)
-```
-
 ## Intended Use
 
 This model is intended for efficient FP8 inference on NVIDIA GPUs with FP8 support (Hopper architecture and above).