Update README.md

README.md

@@ -650,7 +650,7 @@ import torch
 from transformers import Qwen2VLForConditionalGeneration, AutoTokenizer, AutoProcessor
 from qwen_vl_utils import process_vision_info
 
-model_path = "erax/EraX-VL-7B-V1.5"
+model_path = "erax-ai/EraX-VL-7B-V2.0-Preview"
 
 model = Qwen2VLForConditionalGeneration.from_pretrained(
     model_path,
@@ -712,13 +712,13 @@ inputs = inputs.to("cuda")
 # Generation configs
 generation_config =  model.generation_config
 generation_config.do_sample   = True
-generation_config.temperature = 1.0
+generation_config.temperature = 0.01
 generation_config.top_k       = 1
-generation_config.top_p       = 0.9
+generation_config.top_p       = 0.1
 generation_config.min_p       = 0.1
-generation_config.best_of     = 5
+generation_config.best_of     = 1
 generation_config.max_new_tokens     = 2048
-generation_config.repetition_penalty = 1.06
+generation_config.repetition_penalty = 1.05
 
 # Inference
 generated_ids = model.generate(**inputs, generation_config=generation_config)