初始化项目，由ModelHub XC社区提供模型

Model: vastai-ais/Qwen3-4B-Instruct-2507-GPTQ-INT4
Source: Original Platform

config.json

@@ -0,0 +1,62 @@
+{
+  "architectures": [
+    "Qwen3ForCausalLM"
+  ],
+  "attention_bias": false,
+  "attention_dropout": 0.0,
+  "bos_token_id": 151643,
+  "eos_token_id": 151645,
+  "head_dim": 128,
+  "hidden_act": "silu",
+  "hidden_size": 2560,
+  "initializer_range": 0.02,
+  "intermediate_size": 9728,
+  "max_position_embeddings": 262144,
+  "max_window_layers": 36,
+  "model_type": "qwen3",
+  "num_attention_heads": 32,
+  "num_hidden_layers": 36,
+  "num_key_value_heads": 8,
+  "quantization_config": {
+    "act_bits": 16,
+    "act_data_type": "int",
+    "act_dynamic": true,
+    "act_group_size": 128,
+    "act_sym": true,
+    "amp": true,
+    "autoround_version": "0.5.1",
+    "batch_size": 8,
+    "bits": 4,
+    "damp_percent": 0.01,
+    "data_type": "int",
+    "desc_act": false,
+    "enable_minmax_tuning": true,
+    "enable_norm_bias_tuning": false,
+    "enable_quanted_input": true,
+    "gradient_accumulate_steps": 1,
+    "group_size": 128,
+    "iters": 1000,
+    "low_gpu_mem_usage": true,
+    "lr": 0.001,
+    "minmax_lr": 0.001,
+    "nsamples": 512,
+    "quant_method": "gptq",
+    "scale_dtype": "torch.float16",
+    "seqlen": 2048,
+    "super_bits": null,
+    "super_group_size": null,
+    "sym": true,
+    "to_quant_block_names": null,
+    "true_sequential": false
+  },
+  "rms_norm_eps": 1e-06,
+  "rope_scaling": null,
+  "rope_theta": 5000000,
+  "sliding_window": null,
+  "tie_word_embeddings": true,
+  "torch_dtype": "float16",
+  "transformers_version": "4.51.3",
+  "use_cache": true,
+  "use_sliding_window": false,
+  "vocab_size": 151936
+}