Update README.md

README.md

@@ -96,15 +96,6 @@ Safety alignment is another aspect we particularly emphasize. The Web3 investmen
 | | DMind Benchmark | - | - | - | - | - |
 
 
-
-| 模型 | MMLU-Pro (EM) | GPQA-Diamond (Pass@1) | SimpleQA (Correct) | AIME 2024 (Pass@1) | AIME 2025 (Pass@1) | CNMO 2024 (Pass@1) | BFCL_v3 |
-|------|---------------|----------------------|-------------------|-------------------|-------------------|-------------------|---------|
-| **DeepSeek-R1-0528-Qwen3-8B** | 需查找更多信息* | **61.1** | 需查找更多信息* | **86.0** | **76.3** | 需查找更多信息* | 需查找更多信息* |
-| **gpt-oss-20b** | 约85.3%** | **约81.4** | **约6.7*** | **约86.2** | **约68.7** | 无数据 | 需查找更多信息* |
-| **Qwen3-32B** | 需查找更多信息* | **65.6** | 需查找更多信息* | **81.4** | **72.9** | 无数据 | **70.3** |
-| **Qwen3-4B(Thinking)** | **74.0** | **65.8** | 需查找更多信息* | 需查找更多信息* | **81.3** | 需查找更多信息* | **71.2** |
-
-
 ## Application Scenarios
 
 ### 🎯 Edge-Side Web3 Investment Decision Support
@@ -132,11 +123,11 @@ import torch
 
 # Load model (optimized for edge deployment)
 model = AutoModelForCausalLM.from_pretrained(
-    "DMind/DMind2-mini",
+    "Qwen/Qwen3-4B-Thinking-2507",
     torch_dtype=torch.float16,  # Use half precision to save VRAM
     device_map="auto"
 )
-tokenizer = AutoTokenizer.from_pretrained("DMind/DMind2-mini")
+tokenizer = AutoTokenizer.from_pretrained("Qwen/Qwen3-4B-Thinking-2507")
 
 # Investment analysis example
 prompt = """
@@ -151,7 +142,7 @@ Please provide investment advice and risk analysis.
 inputs = tokenizer(prompt, return_tensors="pt")
 outputs = model.generate(
     **inputs, 
-    max_length=2048,
+    max_length=8192,
     temperature=0.7,
     do_sample=True
 )
@@ -187,7 +178,7 @@ This model follows the Apache 2.0 open-source license. Commercial use must compl
 @misc{dmind2024,
   title={DMind-2: Advanced Web3 Domain-Specific Large Language Models with Distribution-Preserving CoT Distillation},
   author={DMind Team},
-  year={2024},
+  year={2025},
   publisher={Hugging Face}
 }
 ```
@@ -195,7 +186,6 @@ This model follows the Apache 2.0 open-source license. Commercial use must compl
 ## Contact
 
 - 🌐 Project Homepage: [https://dmind.ai](https://dmind.ai)
-- 📧 Technical Support: tech@dmind.ai
 - 💬 Community Discussion: [Discord](https://discord.gg/dmind)
 - 🐦 Twitter: [@DMindAI](https://twitter.com/DMindAI)