Gemma 4 E4B it NVFP4 NVFP4 quantized version of google/gemma 4 E4B it for vLLM . Quantization Profile Text backbone: NVFP4 lm head : higher precision Vision tower and vision embeddings: higher precision Audio tower and audio embeddings: higher precision KV cache: FP8 Usage Official Gemma 4 vLLM recipe: https://docs.vllm.ai/projects/recipes/en/latest/Google/Gemma4.html Text Generation
We use cookies for essential functionality and analytics. You can accept or reject analytics cookies.Cookie policy