auto-coder 0.1.255__tar.gz → 0.1.257__tar.gz

{auto_coder-0.1.255 → auto_coder-0.1.257}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: auto-coder
-Version: 0.1.255
+Version: 0.1.257
 Summary: AutoCoder: AutoCoder
 Author: allwefantasy
 Classifier: Topic :: Scientific/Engineering :: Artificial Intelligence
@@ -26,7 +26,7 @@ Requires-Dist: tabulate
 Requires-Dist: jupyter_client
 Requires-Dist: prompt-toolkit
 Requires-Dist: tokenizers
-Requires-Dist: byzerllm[saas]>=0.1.163
+Requires-Dist: byzerllm[saas]>=0.1.165
 Requires-Dist: patch
 Requires-Dist: diff_match_patch
 Requires-Dist: GitPython

{auto_coder-0.1.255 → auto_coder-0.1.257}/src/auto_coder.egg-info/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: auto-coder
-Version: 0.1.255
+Version: 0.1.257
 Summary: AutoCoder: AutoCoder
 Author: allwefantasy
 Classifier: Topic :: Scientific/Engineering :: Artificial Intelligence
@@ -26,7 +26,7 @@ Requires-Dist: tabulate
 Requires-Dist: jupyter_client
 Requires-Dist: prompt-toolkit
 Requires-Dist: tokenizers
-Requires-Dist: byzerllm[saas]>=0.1.163
+Requires-Dist: byzerllm[saas]>=0.1.165
 Requires-Dist: patch
 Requires-Dist: diff_match_patch
 Requires-Dist: GitPython

{auto_coder-0.1.255 → auto_coder-0.1.257}/src/auto_coder.egg-info/SOURCES.txt

@@ -98,6 +98,8 @@ src/autocoder/index/types.py
 src/autocoder/index/filter/__init__.py
 src/autocoder/index/filter/normal_filter.py
 src/autocoder/index/filter/quick_filter.py
+src/autocoder/privacy/__init__.py
+src/autocoder/privacy/model_filter.py
 src/autocoder/pyproject/__init__.py
 src/autocoder/rag/__init__.py
 src/autocoder/rag/api_server.py
@@ -139,6 +141,7 @@ src/autocoder/utils/conversation_store.py
 src/autocoder/utils/llm_client_interceptors.py
 src/autocoder/utils/llms.py
 src/autocoder/utils/log_capture.py
+src/autocoder/utils/model_provider_selector.py
 src/autocoder/utils/multi_turn.py
 src/autocoder/utils/operate_config_api.py
 src/autocoder/utils/print_table.py
@@ -157,5 +160,6 @@ tests/test_chat_auto_coder.py
 tests/test_code_auto_merge_editblock.py
 tests/test_command_completer.py
 tests/test_planner.py
+tests/test_privacy.py
 tests/test_queue_communicate.py
 tests/test_symbols_utils.py

{auto_coder-0.1.255 → auto_coder-0.1.257}/src/auto_coder.egg-info/requires.txt

@@ -16,7 +16,7 @@ tabulate
 jupyter_client
 prompt-toolkit
 tokenizers
-byzerllm[saas]>=0.1.163
+byzerllm[saas]>=0.1.165
 patch
 diff_match_patch
 GitPython

{auto_coder-0.1.255 → auto_coder-0.1.257}/src/autocoder/auto_coder.py

@@ -47,6 +47,8 @@ from autocoder.common.utils_code_auto_generate import stream_chat_with_continue
 from autocoder.utils.auto_coder_utils.chat_stream_out import stream_out
 from autocoder.common.printer import Printer
 from autocoder.rag.token_counter import count_tokens
+from autocoder.privacy.model_filter import ModelPathFilter
+
 console = Console()
 
 
@@ -317,53 +319,7 @@ def main(input_args: Optional[List[str]] = None):
                     "saas.model": model_info["model_name"],
                     "saas.is_reasoning": model_info["is_reasoning"]
                 }
-            )                
-                        
-            if models_module.check_model_exists("deepseek_r1_chat"):
-                r1_model_info = models_module.get_model_by_name("deepseek_r1_chat")  
-                api_key = r1_model_info["api_key"]
-                chat_llm = byzerllm.SimpleByzerLLM(default_model_name="deepseek_r1_chat")
-                chat_llm.deploy(
-                    model_path="",
-                    pretrained_model_type="saas/openai",
-                    udf_name="deepseek_r1_chat",
-                    infer_params={
-                        "saas.base_url": "https://api.deepseek.com/v1",
-                        "saas.api_key": api_key,
-                        "saas.model": "deepseek-reasoner",
-                        "saas.is_reasoning": True
-                    }
-                )
-
-                generate_rerank_llm = byzerllm.SimpleByzerLLM(default_model_name="deepseek_r1_chat")
-                generate_rerank_llm.deploy(
-                    model_path="",
-                    pretrained_model_type="saas/openai",
-                    udf_name="deepseek_r1_chat",
-                    infer_params={
-                        "saas.base_url": "https://api.deepseek.com/v1",
-                        "saas.api_key": api_key,
-                        "saas.model": "deepseek-reasoner",
-                        "saas.is_reasoning": True
-                    }
-                )
-
-                index_filter_llm = byzerllm.SimpleByzerLLM(default_model_name="deepseek_r1_chat")
-                index_filter_llm.deploy(
-                    model_path="",
-                    pretrained_model_type="saas/openai",
-                    udf_name="deepseek_r1_chat",
-                    infer_params={
-                        "saas.base_url": "https://api.deepseek.com/v1",
-                        "saas.api_key": api_key,
-                        "saas.model": "deepseek-reasoner",
-                        "saas.is_reasoning": True
-                    }
-                )
-                                                
-                llm.setup_sub_client("chat_model", chat_llm)
-                llm.setup_sub_client("generate_rerank_model", generate_rerank_llm)
-                llm.setup_sub_client("index_filter_model", index_filter_llm)
+            )                                                    
 
         if args.product_mode == "lite":                                    
             # Set up default models based on configuration
@@ -947,6 +903,17 @@ def main(input_args: Optional[List[str]] = None):
             # )
             return
         elif raw_args.agent_command == "project_reader":
+                        
+            target_llm = llm.get_sub_client("planner_model")
+            if not target_llm:
+                target_llm = llm
+            model_filter = ModelPathFilter.from_model_object(target_llm, args)
+            if model_filter.has_rules():
+                printer = Printer()
+                msg = printer.get_message_from_key_with_format("model_has_access_restrictions",                                            
+                                           model_name=",".join(get_llm_names(target_llm)))
+                raise ValueError(msg)
+
             from autocoder.agent.project_reader import ProjectReader
 
             project_reader = ProjectReader(args, llm)
@@ -1194,10 +1161,23 @@ def main(input_args: Optional[List[str]] = None):
             else:
                 pp = SuffixProject(args=args, llm=llm, file_filter=None)
             pp.run()
-            sources = pp.sources                       
-
+            sources = pp.sources
+            
+            # Apply model filter for chat_llm            
+            model_filter = ModelPathFilter.from_model_object(chat_llm, args)
+            filtered_sources = []
+            printer = Printer()            
+            for source in sources:
+                if model_filter.is_accessible(source.module_name):
+                    filtered_sources.append(source)
+                else:                    
+                    printer.print_in_terminal("index_file_filtered", 
+                                               style="yellow",
+                                               file_path=source.module_name, 
+                                               model_name=",".join(get_llm_names(chat_llm)))
+            
             s = build_index_and_filter_files(
-                llm=llm, args=args, sources=sources)                        
+                llm=llm, args=args, sources=filtered_sources).to_str()                        
             
             if s:
                 pre_conversations.append(
@@ -1389,11 +1369,25 @@ def main(input_args: Optional[List[str]] = None):
                 elapsed_time = time.time() - start_time
                 printer = Printer()
                 speed = last_meta.generated_tokens_count / elapsed_time
+                
+                # Get model info for pricing
+                from autocoder.utils import llms as llm_utils
+                model_info = llm_utils.get_model_info(model_name, args.product_mode) or {}
+                input_price = model_info.get("input_price", 0.0) if model_info else 0.0
+                output_price = model_info.get("output_price", 0.0) if model_info else 0.0
+                
+                # Calculate costs
+                input_cost = (last_meta.input_tokens_count * input_price) / 1000000  # Convert to millions
+                output_cost = (last_meta.generated_tokens_count * output_price) / 1000000  # Convert to millions
+                
                 printer.print_in_terminal("stream_out_stats", 
+                                    model_name=model_name,
                                     elapsed_time=elapsed_time,
                                     first_token_time=last_meta.first_token_time,
                                     input_tokens=last_meta.input_tokens_count,
                                     output_tokens=last_meta.generated_tokens_count,
+                                    input_cost=round(input_cost, 4),
+                                    output_cost=round(output_cost, 4),
                                     speed=round(speed, 2))
             
             chat_history["ask_conversation"].append(

{auto_coder-0.1.255 → auto_coder-0.1.257}/src/autocoder/chat_auto_coder.py

@@ -297,23 +297,22 @@ def initialize_system(args):
 
     init_project()
 
-    if args.product_mode == "lite":
-        # Setup deepseek api key
-        api_key_dir = os.path.expanduser("~/.auto-coder/keys")
-        api_key_file = os.path.join(api_key_dir, "api.deepseek.com")
-        
-        if not os.path.exists(api_key_file):
-            print_status(get_message("model_not_available"), "warning")
-            api_key = prompt(HTML(f"<b>{get_message('enter_api_key')} </b>"))
-            
-            # Create directory if it doesn't exist
-            os.makedirs(api_key_dir, exist_ok=True)
-            
-            # Save the API key
-            with open(api_key_file, "w") as f:
-                f.write(api_key)
-            
-            print_status(f"API key saved successfully: {api_key_file}", "success")                    
+    if args.product_mode == "lite":        
+        from autocoder.utils.model_provider_selector import ModelProviderSelector
+        from autocoder import models as models_module
+        if not models_module.check_model_exists("v3_chat") or not models_module.check_model_exists("r1_chat"):
+            model_provider_selector = ModelProviderSelector()
+            model_provider_info = model_provider_selector.select_provider()
+            if model_provider_info is not None:
+                models_json_list = model_provider_selector.to_models_json(model_provider_info)
+                models_module.add_and_activate_models(models_json_list)
+                r1_model = models_json_list[0]['name']
+                v3_model = models_json_list[1]['name']            
+                configure(f"model:{v3_model}", skip_print=True)
+                configure(f"chat_model:{r1_model}", skip_print=True)
+                configure(f"generate_rerank_model:{r1_model}", skip_print=True)
+                configure(f"code_model:{v3_model}", skip_print=True)
+                configure(f"index_filter_model:{r1_model}", skip_print=True)            
 
     if args.product_mode == "pro":
         # Check if Ray is running

{auto_coder-0.1.255 → auto_coder-0.1.257}/src/autocoder/chat_auto_coder_lang.py

@@ -128,7 +128,7 @@ MESSAGES = {
         "official_doc": "Official Documentation: https://uelng8wukz.feishu.cn/wiki/NhPNwSRcWimKFIkQINIckloBncI",
     },
     "zh": {
-        "commit_generating": "{{ model_name }} 正在生成提交信息...", 
+        "commit_generating": "{{ model_name }} 正在生成提交信息...",
         "commit_message": "{{ model_name }} 生成的提交信息: {{ message }}",
         "commit_failed": "{{ model_name }} 生成提交信息失败: {{ error }}",
         "mcp_remove_error": "移除 MCP 服务器时出错:{error}",

{auto_coder-0.1.255 → auto_coder-0.1.257}/src/autocoder/common/__init__.py

@@ -14,7 +14,13 @@ class SourceCode(pydantic.BaseModel):
     tokens: int = -1
     metadata: Dict[str, Any] = {}
 
+class SourceCodeList():
+    def __init__(self, sources: List[SourceCode]):
+        self.sources = sources
 
+    def to_str(self):
+        return "\n".join([f"##File: {source.module_name}\n{source.source_code}\n" for source in self.sources])
+    
 class TranslateReadme(pydantic.BaseModel):
     filename: str = pydantic.Field(..., description="需要翻译的文件路径")
     content: str = pydantic.Field(..., description="翻译后的内容")
@@ -362,6 +368,7 @@ class AutoCoderArgs(pydantic.BaseModel):
     keep_only_reasoning_content: Optional[bool] = False
 
     in_code_apply: bool = False
+    model_filter_path: Optional[str] = None
 
     class Config:
         protected_namespaces = ()

{auto_coder-0.1.255 → auto_coder-0.1.257}/src/autocoder/common/auto_coder_lang.py

@@ -3,6 +3,20 @@ from byzerllm.utils import format_str_jinja2
 
 MESSAGES = {
     "en": {
+        "model_provider_select_title": "Select Model Provider",
+        "model_provider_select_text": "Please select your model provider:",
+        "model_provider_volcano": "Volcano Engine",
+        "model_provider_guiji": "GuiJi AI",
+        "model_provider_deepseek": "DeepSeek Official",        
+        "model_provider_api_key_title": "API Key",
+        "model_provider_volcano_api_key_text": "Please enter your Volcano Engine API key:",
+        "model_provider_volcano_r1_text": "Please enter your Volcano Engine R1 endpoint (format: ep-20250204215011-vzbsg):",
+        "model_provider_volcano_v3_text": "Please enter your Volcano Engine V3 endpoint (format: ep-20250204215011-vzbsg):",
+        "model_provider_guiji_api_key_text": "Please enter your GuiJi AI API key:",
+        "model_provider_deepseek_api_key_text": "Please enter your DeepSeek API key:",
+        "model_provider_selected": "Provider configuration completed successfully! You can use /models command to view, add and modify all models later.",
+        "model_provider_success_title": "Success",
+        "index_file_filtered": "File {{file_path}} is filtered by model {{model_name}} restrictions",
         "models_no_active": "No active models found",
         "models_speed_test_results": "Model Speed Test Results",
         "models_testing": "Testing model: {{name}}...",
@@ -11,7 +25,7 @@ MESSAGES = {
         "generation_cancelled": "[Interrupted] Generation cancelled",
         "model_not_found": "Model {{model_name}} not found",
         "generating_shell_script": "Generating Shell Script",
-        "new_session_started": "New session started. Previous chat history has been archived.",    
+        "new_session_started": "New session started. Previous chat history has been archived.",
         "memory_save_success": "✅ Saved to your memory",
         "file_decode_error": "Failed to decode file: {{file_path}}. Tried encodings: {{encodings}}",
         "file_write_error": "Failed to write file: {{file_path}}. Error: {{error}}",
@@ -21,7 +35,7 @@ MESSAGES = {
         "no_latest_commit": "Unable to get latest commit information",
         "code_review_error": "Code review process error: {{error}}",
         "index_file_too_large": "⚠️ File {{ file_path }} is too large ({{ file_size }} > {{ max_length }}), splitting into chunks...",
-        "index_update_success": "✅ {{ model_name }} Successfully updated index for {{ file_path }} (md5: {{ md5 }}) in {{ duration }}s",
+        "index_update_success": "✅ {{ model_name }} Successfully updated index for {{ file_path }} (md5: {{ md5 }}) in {{ duration }}s, input_tokens: {{ input_tokens }}, output_tokens: {{ output_tokens }}, input_cost: {{ input_cost }}, output_cost: {{ output_cost }}",
         "index_build_error": "❌ {{ model_name }} Error building index for {{ file_path }}: {{ error }}",
         "index_build_summary": "📊 Total Files: {{ total_files }}, Need to Build Index: {{ num_files }}",
         "building_index_progress": "⏳ Building Index: {{ counter }}/{{ num_files }}...",
@@ -30,7 +44,7 @@ MESSAGES = {
         "index_threads_completed": "✅ Completed {{ completed_threads }}/{{ total_threads }} threads",
         "index_related_files_fail": "⚠️ Failed to find related files for chunk {{ chunk_count }}",
         "index_file_removed": "🗑️ Removed non-existent file index: {{ file_path }}",
-        "index_file_saved": "💾 Saved index file, updated {{ updated_files }} files, removed {{ removed_files }} files",
+        "index_file_saved": "💾 Saved index file, updated {{ updated_files }} files, removed {{ removed_files }} files, input_tokens: {{ input_tokens }}, output_tokens: {{ output_tokens }}, input_cost: {{ input_cost }}, output_cost: {{ output_cost }}",
         "human_as_model_instructions": (
             "You are now in Human as Model mode. The content has been copied to your clipboard.\n"
             "The system is waiting for your input. When finished, enter 'EOF' on a new line to submit.\n"
@@ -57,7 +71,7 @@ MESSAGES = {
             "Paste the answer to the input box below, use '/break' to exit, '/clear' to clear the screen, '/eof' to submit."
         ),
         "code_generation_start": "Auto generate the code...",
-        "code_generation_complete": "Code generation completed in {{ duration }} seconds, input_tokens_count: {{ input_tokens }}, generated_tokens_count: {{ output_tokens }}, speed: {{ speed }} tokens/s",
+        "code_generation_complete": "{{ model_names}} Code generation completed in {{ duration }} seconds, input_tokens_count: {{ input_tokens }}, generated_tokens_count: {{ output_tokens }}, input_cost: {{ input_cost }}, output_cost: {{ output_cost }}, speed: {{ speed }} tokens/s",
         "code_merge_start": "Auto merge the code...",
         "code_execution_warning": "Content(send to model) is {{ content_length }} tokens (you may collect too much files), which is larger than the maximum input length {{ max_length }}",
         "quick_filter_start": "{{ model_name }} Starting filter context(quick_filter)...",
@@ -75,13 +89,13 @@ MESSAGES = {
         "ranking_start": "Start ranking {{ count }} candidates using model {{ model_name }}",
         "ranking_failed_request": "Ranking request failed: {{ error }}",
         "ranking_all_failed": "All ranking requests failed",
-        "ranking_complete": "Ranking completed in {{ elapsed }}s, total voters: {{ total_tasks }}, best candidate index: {{ best_candidate }}, scores: {{ scores }}, input_tokens: {{ input_tokens }}, output_tokens: {{ output_tokens }}",
+        "ranking_complete": "{{ model_names }} Ranking completed in {{ elapsed }}s, total voters: {{ total_tasks }}, best candidate index: {{ best_candidate }}, scores: {{ scores }}, input_tokens: {{ input_tokens }}, output_tokens: {{ output_tokens }}, input_cost: {{ input_cost }}, output_cost: {{ output_cost }}",
         "ranking_process_failed": "Ranking process failed: {{ error }}",
         "ranking_failed": "Ranking failed in {{ elapsed }}s, using original order",
         "begin_index_source_code": "🚀 Begin to index source code in {{ source_dir }}",
-        "stream_out_stats": "Elapsed time {{ elapsed_time }} seconds, first token time: {{ first_token_time }} seconds, input tokens: {{ input_tokens }}, output tokens: {{ output_tokens }}, speed: {{ speed }} tokens/s",
-        "quick_filter_stats": "快速过滤器完成，耗时 {{ elapsed_time }} 秒，输入token数: {{ input_tokens }}, 输出token数: {{ output_tokens }}",
-        "upsert_file": "✅ Updated file: {{ file_path }}",  
+        "stream_out_stats": "Model: {{ model_name }}, Total time: {{ elapsed_time }} seconds, First token time: {{ first_token_time }} seconds, Speed: {{ speed }} tokens/s, Input tokens: {{ input_tokens }}, Output tokens: {{ output_tokens }}, Input cost: {{ input_cost }}, Output cost: {{ output_cost }}",
+        "quick_filter_stats": "{{ model_names }} 快速过滤器完成，耗时 {{ elapsed_time }} 秒，输入token数: {{ input_tokens }}, 输出token数: {{ output_tokens }}, 输入成本: {{ input_cost }}, 输出成本: {{ output_cost }}",
+        "upsert_file": "✅ Updated file: {{ file_path }}",
         "unmerged_blocks_title": "Unmerged Blocks",
         "quick_filter_title": "{{ model_name }} is analyzing how to filter context...",
         "quick_filter_failed": "❌ Quick filter failed: {{ error }}. ",
@@ -95,8 +109,23 @@ MESSAGES = {
         "quick_filter_tokens_len": "📊 Current index size: {{ tokens_len }} tokens",
         "estimated_chat_input_tokens": "Estimated chat input tokens: {{ estimated_input_tokens }}",
         "estimated_input_tokens_in_generate": "Estimated input tokens in generate ({{ generate_mode }}): {{ estimated_input_tokens }}",
+        "model_has_access_restrictions": "{{model_name}} has access restrictions, cannot use the current function",
     },
     "zh": {
+        "model_provider_select_title": "选择模型供应商",
+        "model_provider_select_text": "请选择您的模型供应商：",
+        "model_provider_volcano": "火山方舟",
+        "model_provider_guiji": "硅基流动",
+        "model_provider_deepseek": "DeepSeek官方",        
+        "model_provider_api_key_title": "API密钥",
+        "model_provider_volcano_api_key_text": "请输入您的火山方舟API密钥：",
+        "model_provider_volcano_r1_text": "请输入您的火山方舟 R1 推理点(格式如: ep-20250204215011-vzbsg)：",
+        "model_provider_volcano_v3_text": "请输入您的火山方舟 V3 推理点(格式如: ep-20250204215011-vzbsg)：",
+        "model_provider_guiji_api_key_text": "请输入您的硅基流动API密钥：",
+        "model_provider_deepseek_api_key_text": "请输入您的DeepSeek API密钥：",
+        "model_provider_selected": "供应商配置已成功完成！后续你可以使用 /models 命令，查看，新增和修改所有模型",
+        "model_provider_success_title": "成功",
+        "index_file_filtered": "文件 {{file_path}} 被模型 {{model_name}} 的访问限制过滤",
         "models_no_active": "未找到激活的模型",
         "models_speed_test_results": "模型速度测试结果",
         "models_testing": "正在测试模型: {{name}}...",
@@ -114,7 +143,7 @@ MESSAGES = {
         "no_latest_commit": "无法获取最新的提交信息",
         "code_review_error": "代码审查过程出错: {{error}}",
         "index_file_too_large": "⚠️ 文件 {{ file_path }} 过大 ({{ file_size }} > {{ max_length }}), 正在分块处理...",
-        "index_update_success": "✅ {{ model_name }} 成功更新 {{ file_path }} 的索引 (md5: {{ md5 }}), 耗时 {{ duration }} 秒",
+        "index_update_success": "✅ {{ model_name }} 成功更新 {{ file_path }} 的索引 (md5: {{ md5 }}), 耗时 {{ duration }} 秒, 输入token数: {{ input_tokens }}, 输出token数: {{ output_tokens }}, 输入成本: {{ input_cost }}, 输出成本: {{ output_cost }}",
         "index_build_error": "❌ {{ model_name }} 构建 {{ file_path }} 索引时出错: {{ error }}",
         "index_build_summary": "📊 总文件数: {{ total_files }}, 需要构建索引: {{ num_files }}",
         "building_index_progress": "⏳ 正在构建索引: {{ counter }}/{{ num_files }}...",
@@ -123,7 +152,7 @@ MESSAGES = {
         "index_threads_completed": "✅ 已完成 {{ completed_threads }}/{{ total_threads }} 个线程",
         "index_related_files_fail": "⚠️ 无法为块 {{ chunk_count }} 找到相关文件",
         "index_file_removed": "🗑️ 已移除不存在的文件索引：{{ file_path }}",
-        "index_file_saved": "💾 已保存索引文件，更新了 {{ updated_files }} 个文件，移除了 {{ removed_files }} 个文件",
+        "index_file_saved": "💾 已保存索引文件，更新了 {{ updated_files }} 个文件，移除了 {{ removed_files }} 个文件，输入token数: {{ input_tokens }}, 输出token数: {{ output_tokens }}, 输入成本: {{ input_cost }}, 输出成本: {{ output_cost }}",
         "human_as_model_instructions": (
             "您现在处于人类作为模型模式。内容已复制到您的剪贴板。\n"
             "系统正在等待您的输入。完成后，在新行输入'EOF'提交。\n"
@@ -150,7 +179,7 @@ MESSAGES = {
             "将获得答案黏贴到下面的输入框，换行后，使用 '/break' 退出，'/clear' 清屏，'/eof' 提交。"
         ),
         "code_generation_start": "正在自动生成代码...",
-        "code_generation_complete": "代码生成完成，耗时 {{ duration }} 秒，输入token数: {{ input_tokens }}, 输出token数: {{ output_tokens }}, 速度: {{ speed }} tokens/秒",
+        "code_generation_complete": "{{ model_names}} 代码生成完成，耗时 {{ duration }} 秒，输入token数: {{ input_tokens }}, 输出token数: {{ output_tokens }}, 输入成本: {{ input_cost }}, 输出成本: {{ output_cost }}, 速度: {{ speed }} tokens/秒",
         "code_merge_start": "正在自动合并代码...",
         "code_execution_warning": "发送给模型的内容长度为 {{ content_length }} tokens（您可能收集了太多文件），超过了最大输入长度 {{ max_length }}",
         "quick_filter_start": "{{ model_name }} 开始查找上下文(quick_filter)...",
@@ -179,17 +208,17 @@ MESSAGES = {
         "ranking_start": "开始对 {{ count }} 个候选项进行排序,使用模型 {{ model_name }} 打分",
         "ranking_failed_request": "排序请求失败: {{ error }}",
         "ranking_all_failed": "所有排序请求都失败",
-        "ranking_complete": "排序完成，耗时 {{ elapsed }} 秒，总投票数: {{ total_tasks }}，最佳候选索引: {{ best_candidate }}，得分: {{ scores }}，输入token数: {{ input_tokens }}，输出token数: {{ output_tokens }}",
+        "ranking_complete": "{{ model_names }} 排序完成，耗时 {{ elapsed }} 秒，总投票数: {{ total_tasks }}，最佳候选索引: {{ best_candidate }}，得分: {{ scores }}，输入token数: {{ input_tokens }}，输出token数: {{ output_tokens }}  输入成本: {{ input_cost }}, 输出成本: {{ output_cost }}",
         "ranking_process_failed": "排序过程失败: {{ error }}",
         "ranking_failed": "排序失败，耗时 {{ elapsed }} 秒，使用原始顺序",
-        "stream_out_stats": "总耗时 {{ elapsed_time }} 秒，首token时间: {{ first_token_time }} 秒，输入token数: {{ input_tokens }}, 输出token数: {{ output_tokens }}, 速度: {{ speed }} tokens/秒",
-        "quick_filter_stats": "Quick filter completed in {{ elapsed_time }} seconds, input tokens: {{ input_tokens }}, output tokens: {{ output_tokens }}",
+        "stream_out_stats": "模型: {{ model_name }},总耗时 {{ elapsed_time }} 秒,首token时间: {{ first_token_time }} 秒, 速度: {{ speed }} tokens/秒, 输入token数: {{ input_tokens }}, 输出token数: {{ output_tokens }}, 输入成本: {{ input_cost }}, 输出成本: {{ output_cost }}",
+        "quick_filter_stats": "{{ model_names }} Quick filter completed in {{ elapsed_time }} seconds, input tokens: {{ input_tokens }}, output tokens: {{ output_tokens }}, input cost: {{ input_cost }}, output cost: {{ output_cost }}",
         "quick_filter_title": "{{ model_name }} 正在分析如何筛选上下文...",
         "quick_filter_failed": "❌ 快速过滤器失败: {{ error }}. ",
         "estimated_chat_input_tokens": "对话输入token预估为: {{ estimated_input_tokens }}",
         "estimated_input_tokens_in_generate": "生成代码({{ generate_mode }})预计输入token数: {{ estimated_input_tokens_in_generate }}",
-    },
-}
+        "model_has_access_restrictions": "{{model_name}} 有访问限制，无法使用当前功能",
+    }}
 
 
 def get_system_language():
@@ -203,5 +232,6 @@ def get_message(key):
     lang = get_system_language()
     return MESSAGES.get(lang, MESSAGES['en']).get(key, MESSAGES['en'][key])
 
+
 def get_message_with_format(msg_key: str, **kwargs):
     return format_str_jinja2(get_message(msg_key), **kwargs)

{auto_coder-0.1.255 → auto_coder-0.1.257}/src/autocoder/common/code_auto_generate.py

@@ -10,6 +10,9 @@ from autocoder.common.utils_code_auto_generate import chat_with_continue
 import json
 from autocoder.common.printer import Printer
 from autocoder.rag.token_counter import count_tokens
+from autocoder.utils import llms as llm_utils
+from autocoder.common import SourceCodeList
+from autocoder.privacy.model_filter import ModelPathFilter
 
 
 class CodeAutoGenerate:
@@ -155,10 +158,27 @@ class CodeAutoGenerate:
         }
 
     def single_round_run(
-        self, query: str, source_content: str
+        self, query: str, source_code_list: SourceCodeList
     ) -> Tuple[List[str], Dict[str, str]]:
         llm_config = {"human_as_model": self.args.human_as_model}
 
+        # Apply model filter for code_llm  
+        printer = Printer()      
+        for llm in self.llms:
+            model_filter = ModelPathFilter.from_model_object(llm, self.args)
+            filtered_sources = []
+            for source in source_code_list.sources:
+                if model_filter.is_accessible(source.module_name):
+                    filtered_sources.append(source)
+                else:
+                    printer.print_in_terminal("index_file_filtered", 
+                                               style="yellow",
+                                               file_path=source.module_name, 
+                                               model_name=",".join(llm_utils.get_llm_names(llm)))
+            
+        source_code_list = SourceCodeList(filtered_sources)
+        source_content = source_code_list.to_str()
+
         if self.args.request_id and not self.args.skip_events:
             queue_communicate.send_event_no_wait(
                 request_id=self.args.request_id,
@@ -193,6 +213,9 @@ class CodeAutoGenerate:
         results = []
         input_tokens_count = 0
         generated_tokens_count = 0
+        input_tokens_cost = 0
+        generated_tokens_cost = 0
+        model_names = []
 
         printer = Printer()
         estimated_input_tokens = count_tokens(json.dumps(conversations, ensure_ascii=False))
@@ -206,13 +229,27 @@ class CodeAutoGenerate:
                 futures = []
                 for llm in self.llms:
                     for _ in range(self.generate_times_same_model):
-                        futures.append(executor.submit(
-                            chat_with_continue, llm=llm, conversations=conversations, llm_config=llm_config))
+                        
+                        model_names_list = llm_utils.get_llm_names(llm)
+                        model_name = None
+                        if model_names_list:
+                            model_name = model_names_list[0]                                                    
+                        
+                        for _ in range(self.generate_times_same_model):
+                            model_names.append(model_name)
+                            futures.append(executor.submit(
+                                chat_with_continue, llm=llm, conversations=conversations, llm_config=llm_config))
+                            
                 temp_results = [future.result() for future in futures]
                 for result in temp_results:
                     results.append(result.content)
                     input_tokens_count += result.input_tokens_count
                     generated_tokens_count += result.generated_tokens_count
+                    model_info = llm_utils.get_model_info(model_name, self.args.product_mode)
+                    input_cost = model_info.get("input_price", 0) if model_info else 0
+                    output_cost = model_info.get("output_price", 0) if model_info else 0
+                    input_tokens_cost += input_cost * result.input_tokens_count / 1000000
+                    generated_tokens_cost += output_cost * result.generated_tokens_count / 1000000
             
             for result in results:
                 conversations_list.append(
@@ -227,7 +264,9 @@ class CodeAutoGenerate:
         
         statistics = {
             "input_tokens_count": input_tokens_count,
-            "generated_tokens_count": generated_tokens_count
+            "generated_tokens_count": generated_tokens_count,
+            "input_tokens_cost": input_tokens_cost,
+            "generated_tokens_cost": generated_tokens_cost
         }        
     
         if self.args.request_id and not self.args.skip_events:
@@ -242,10 +281,11 @@ class CodeAutoGenerate:
         return CodeGenerateResult(contents=results, conversations=conversations_list, metadata=statistics)
 
     def multi_round_run(
-        self, query: str, source_content: str, max_steps: int = 10
+        self, query: str, source_code_list: SourceCodeList, max_steps: int = 10
     ) -> Tuple[List[str], List[Dict[str, str]]]:
         llm_config = {"human_as_model": self.args.human_as_model}
         result = []
+        source_content = source_code_list.to_str()
 
         if self.args.template == "common":
             init_prompt = self.multi_round_instruction.prompt(