auto-coder 0.1.250__py3-none-any.whl → 0.1.252__py3-none-any.whl

autocoder/chat_auto_coder.py

@@ -54,7 +54,7 @@ import shlex
 from autocoder.utils.llms import get_single_llm
 import pkg_resources
 from autocoder.common.printer import Printer
-from byzerllm.utils.langutil import run_in_thread,run_in_raw_thread
+from autocoder.utils.thread_utils import run_in_thread,run_in_raw_thread
 
 class SymbolItem(BaseModel):
     symbol_name: str
@@ -136,6 +136,8 @@ commands = [
 
 
 def show_help():
+    print(f"\033[1m{get_message('official_doc')}\033[0m")
+    print()
     print(f"\033[1m{get_message('supported_commands')}\033[0m")
     print()
     print(
@@ -1713,7 +1715,7 @@ def commit(query: str):
                 if os.path.exists(temp_yaml):
                     os.remove(temp_yaml)
             
-            target_model = args.code_model or args.model
+            target_model = args.commit_model or args.model
             llm = get_single_llm(target_model, product_mode)
             printer = Printer()
             printer.print_in_terminal("commit_generating", style="yellow", model_name=target_model)
@@ -2208,6 +2210,9 @@ def manage_models(params, query: str):
             table.add_column("Name", style="cyan", width=40, no_wrap=False)
             table.add_column("Model Name", style="magenta", width=30, overflow="fold")
             table.add_column("Base URL", style="white", width=50, overflow="fold")
+            table.add_column("Input Price (M)", style="magenta", width=15)
+            table.add_column("Output Price (M)", style="magenta", width=15)
+            table.add_column("Speed (s/req)", style="blue", width=15)
             for m in models_data:
                 # Check if api_key_path exists and file exists
                 is_api_key_set = "api_key" in m  
@@ -2221,12 +2226,60 @@ def manage_models(params, query: str):
                 table.add_row(
                     name,                    
                     m.get("model_name", ""),                    
-                    m.get("base_url", "")
+                    m.get("base_url", ""),
+                    f"{m.get('input_price', 0.0):.2f}",
+                    f"{m.get('output_price', 0.0):.2f}",
+                    f"{m.get('average_speed', 0.0):.3f}"
                 )
             console.print(table)
         else:
             printer.print_in_terminal("models_no_models", style="yellow")
 
+    elif subcmd == "/input_price":
+        args = query.strip().split()
+        if len(args) >= 2:
+            name = args[0]
+            try:
+                price = float(args[1])
+                if models.update_model_input_price(name, price):
+                    printer.print_in_terminal("models_input_price_updated", style="green", name=name, price=price)
+                else:
+                    printer.print_in_terminal("models_not_found", style="red", name=name)
+            except ValueError as e:
+                printer.print_in_terminal("models_invalid_price", style="red", error=str(e))
+        else:
+            printer.print_in_terminal("models_input_price_usage", style="red")
+            
+    elif subcmd == "/output_price":
+        args = query.strip().split()
+        if len(args) >= 2:
+            name = args[0]
+            try:
+                price = float(args[1])
+                if models.update_model_output_price(name, price):
+                    printer.print_in_terminal("models_output_price_updated", style="green", name=name, price=price)
+                else:
+                    printer.print_in_terminal("models_not_found", style="red", name=name)
+            except ValueError as e:
+                printer.print_in_terminal("models_invalid_price", style="red", error=str(e))
+        else:
+            printer.print_in_terminal("models_output_price_usage", style="red")
+
+    elif subcmd == "/speed":
+        args = query.strip().split()
+        if len(args) >= 2:
+            name = args[0]
+            try:
+                speed = float(args[1])
+                if models.update_model_speed(name, speed):
+                    printer.print_in_terminal("models_speed_updated", style="green", name=name, speed=speed)
+                else:
+                    printer.print_in_terminal("models_not_found", style="red", name=name)
+            except ValueError as e:
+                printer.print_in_terminal("models_invalid_speed", style="red", error=str(e))
+        else:
+            printer.print_in_terminal("models_speed_usage", style="red")
+            
     elif subcmd == "/add":
         # Support both simplified and legacy formats
         args = query.strip().split(" ")        
@@ -2796,8 +2849,8 @@ def main():
                     command = user_input[len("/shell"):].strip()
                     if not command:
                         print("Please enter a shell command to execute.")
-                else:
-                    execute_shell_command(command)
+                    else:
+                        execute_shell_command(command)
 
         except KeyboardInterrupt:
             continue

autocoder/chat_auto_coder_lang.py

@@ -85,7 +85,7 @@ MESSAGES = {
         "design_desc": "Generate SVG image based on the provided description",
         "commit_desc": "Auto generate yaml file and commit changes based on user's manual changes",
         "models_desc": "Manage model configurations, only available in lite mode",
-        "models_usage": "Usage: /models /list|/add|/add_model|/remove ...",
+        "models_usage": "Usage: /models /list|/add|/add_model|/remove|/price|/speed ...",
         "models_added": "Added/Updated model '{{name}}' successfully.",
         "models_add_failed": "Failed to add model '{{name}}'. Model not found in defaults.",
         "models_add_usage": "Usage: /models /add <name> <api_key> or\n/models /add <name> <model_type> <model_name> <base_url> <api_key_path> [description]",
@@ -96,6 +96,14 @@ MESSAGES = {
         "models_add_model_remove": "Model '{{name}}' not found.",
         "models_add_model_removed": "Removed model: {{name}}",
         "models_unknown_subcmd": "Unknown subcommand: {{subcmd}}",
+        "models_input_price_updated": "Updated input price for model {{name}} to {{price}} M/token",
+        "models_output_price_updated": "Updated output price for model {{name}} to {{price}} M/token",
+        "models_invalid_price": "Invalid price value: {{error}}",
+        "models_input_price_usage": "Usage: /models /input_price <name> <value>",
+        "models_output_price_usage": "Usage: /models /output_price <name> <value>",
+        "models_speed_updated": "Updated speed for model {{name}} to {{speed}} s/request",
+        "models_invalid_speed": "Invalid speed value: {{error}}",
+        "models_speed_usage": "Usage: /models /speed <name> <value>",
         "models_title": "All Models (内置 + models.json)",
         "models_no_models": "No models found.",
         "models_lite_only": "The /models command is only available in lite mode",
@@ -117,6 +125,7 @@ MESSAGES = {
         "commit_message": "{{ model_name }} Generated commit message: {{ message }}",
         "commit_failed": "{{ model_name }} Failed to generate commit message: {{ error }}",
         "confirm_execute": "Do you want to execute this script?",
+        "official_doc": "Official Documentation: https://uelng8wukz.feishu.cn/wiki/NhPNwSRcWimKFIkQINIckloBncI",
     },
     "zh": {
         "commit_generating": "{{ model_name }} 正在生成提交信息...", 
@@ -204,7 +213,7 @@ MESSAGES = {
         "conf_value": "值",
         "conf_title": "配置设置",
         "conf_subtitle": "使用 /conf <key>:<value> 修改这些设置",
-        "models_usage": "用法: /models /list|/add|/add_model|/remove ...",
+        "models_usage": "用法: /models /list|/add|/add_model|/remove|/price|/speed ...",
         "models_added": "成功添加/更新模型 '{{name}}'。",
         "models_add_failed": "添加模型 '{{name}}' 失败。在默认模型中未找到该模型。",
         "models_add_usage": "用法: /models /add <name> <api_key> 或\n/models /add <name> <model_type> <model_name> <base_url> <api_key_path> [description]",
@@ -215,6 +224,14 @@ MESSAGES = {
         "models_add_model_remove": "找不到模型 '{{name}}'。",
         "models_add_model_removed": "已移除模型: {{name}}",
         "models_unknown_subcmd": "未知的子命令: {{subcmd}}",
+        "models_input_price_updated": "已更新模型 {{name}} 的输入价格为 {{price}} M/token",
+        "models_output_price_updated": "已更新模型 {{name}} 的输出价格为 {{price}} M/token",
+        "models_invalid_price": "无效的价格值: {{error}}",
+        "models_input_price_usage": "用法: /models /input_price <name> <value>",
+        "models_output_price_usage": "用法: /models /output_price <name> <value>",
+        "models_speed_updated": "已更新模型 {{name}} 的速度为 {{speed}} 秒/请求",
+        "models_invalid_speed": "无效的速度值: {{error}}",
+        "models_speed_usage": "用法: /models /speed <name> <value>",
         "models_title": "所有模型 (内置 + models.json)",
         "models_no_models": "未找到任何模型。",
         "models_lite_only": "/models 命令仅在 lite 模式下可用",
@@ -232,7 +249,8 @@ MESSAGES = {
         "remove_files_none": "没有文件被移除。",
         "files_removed": "移除的文件",
         "models_api_key_empty": "警告:  {{name}}  API key 为空。请设置一个有效的 API key。",
-        "confirm_execute": "是否执行此脚本？",
+        "confirm_execute": "是否执行此脚本?",
+        "official_doc": "官方文档: https://uelng8wukz.feishu.cn/wiki/NhPNwSRcWimKFIkQINIckloBncI",
     }
 }
 

autocoder/common/__init__.py

@@ -254,6 +254,7 @@ class AutoCoderArgs(pydantic.BaseModel):
     planner_model: Optional[str] = ""
     voice2text_model: Optional[str] = ""
     text2voice_model: Optional[str] = ""
+    commit_model: Optional[str] = ""
     
     skip_build_index: Optional[bool] = False
     skip_filter_index: Optional[bool] = False
@@ -357,7 +358,7 @@ class AutoCoderArgs(pydantic.BaseModel):
     enable_global_memory: Optional[bool] = True  
     product_mode: Optional[str] = "lite"
 
-    keep_reasoning_content: Optional[bool] = True
+    keep_reasoning_content: Optional[bool] = False
 
     in_code_apply: bool = False
 

autocoder/common/auto_coder_lang.py

@@ -3,6 +3,7 @@ from byzerllm.utils import format_str_jinja2
 
 MESSAGES = {
     "en": {
+        "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.",    
@@ -51,7 +52,7 @@ 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 }}",
+        "code_generation_complete": "Code generation completed in {{ duration }} seconds, input_tokens_count: {{ input_tokens }}, generated_tokens_count: {{ output_tokens }}, 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)...",
@@ -73,7 +74,7 @@ MESSAGES = {
         "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, input tokens: {{ input_tokens }}, output tokens: {{ output_tokens }}",
+        "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 }}",  
         "unmerged_blocks_title": "Unmerged Blocks",
@@ -86,9 +87,12 @@ MESSAGES = {
         "git_init_required": "⚠️ auto_merge only applies to git repositories.\n\nPlease try using git init in the source directory:\n\n```shell\ncd {{ source_dir }}\ngit init.\n```\n\nThen run auto - coder again.\nError: {{ error }}",
         "quick_filter_reason": "Auto get(quick_filter mode)",
         "quick_filter_too_long": "⚠️ index file is too large ({{ tokens_len }}/{{ max_tokens }}). The query will be split into {{ split_size }} chunks.",
-        "quick_filter_tokens_len": "📊 Current index size: {{ tokens_len }} tokens"
+        "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 }}",
     },
     "zh": {
+        "generation_cancelled": "[已中断] 生成已取消",
         "model_not_found": "未找到模型: {{model_name}}",
         "generating_shell_script": "正在生成 Shell 脚本",
         "new_session_started": "新会话已开始。之前的聊天历史已存档。",
@@ -137,7 +141,7 @@ 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 }}",
+        "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_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)...",
@@ -169,10 +173,12 @@ MESSAGES = {
         "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_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": "耗时 {{ elapsed_time }} 秒，输入token数: {{ input_tokens }}, 输出token数: {{ output_tokens }}",
+        "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 }}",
         "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 }}",
     },
 }
 

autocoder/common/code_auto_generate.py

@@ -8,6 +8,8 @@ from concurrent.futures import ThreadPoolExecutor
 from autocoder.common.types import CodeGenerateResult
 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
 
 
 class CodeAutoGenerate:
@@ -191,6 +193,14 @@ class CodeAutoGenerate:
         results = []
         input_tokens_count = 0
         generated_tokens_count = 0
+
+        printer = Printer()
+        estimated_input_tokens = count_tokens(json.dumps(conversations, ensure_ascii=False))
+        printer.print_in_terminal("estimated_input_tokens_in_generate", style="yellow",
+                                  estimated_input_tokens_in_generate=estimated_input_tokens,
+                                  generate_mode="wholefile"
+                                  )
+
         if not self.args.human_as_model:
             with ThreadPoolExecutor(max_workers=len(self.llms) * self.generate_times_same_model) as executor:
                 futures = []

autocoder/common/code_auto_generate_diff.py

@@ -7,6 +7,8 @@ from autocoder.common import sys_prompt
 from concurrent.futures import ThreadPoolExecutor
 import json
 from autocoder.common.utils_code_auto_generate import chat_with_continue
+from autocoder.common.printer import Printer
+from autocoder.rag.token_counter import count_tokens
 
 
 class CodeAutoGenerateDiff:
@@ -339,6 +341,14 @@ class CodeAutoGenerateDiff:
         results = []
         input_tokens_count = 0
         generated_tokens_count = 0
+
+        printer = Printer()
+        estimated_input_tokens = count_tokens(json.dumps(conversations, ensure_ascii=False))
+        printer.print_in_terminal("estimated_input_tokens_in_generate", style="yellow",
+                                  estimated_input_tokens_in_generate=estimated_input_tokens,
+                                  generate_mode="diff"
+                                  )
+
         if not self.args.human_as_model:
             with ThreadPoolExecutor(max_workers=len(self.llms) * self.generate_times_same_model) as executor:
                 futures = []

autocoder/common/code_auto_generate_editblock.py

@@ -11,6 +11,8 @@ from autocoder.utils.queue_communicate import (
 import json
 from concurrent.futures import ThreadPoolExecutor
 from autocoder.common.utils_code_auto_generate import chat_with_continue
+from autocoder.common.printer import Printer
+from autocoder.rag.token_counter import count_tokens
 
 
 class CodeAutoGenerateEditBlock:
@@ -421,35 +423,47 @@ class CodeAutoGenerateEditBlock:
         results = []
         input_tokens_count = 0
         generated_tokens_count = 0
+
+        printer = Printer()
+        estimated_input_tokens = count_tokens(
+            json.dumps(conversations, ensure_ascii=False))
+        printer.print_in_terminal("estimated_input_tokens_in_generate",
+                                  style="yellow",
+                                  estimated_input_tokens_in_generate=estimated_input_tokens,
+                                  generate_mode="editblock"
+                                  )
+
         if not self.args.human_as_model:
             with ThreadPoolExecutor(max_workers=len(self.llms) * self.generate_times_same_model) as executor:
                 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))
+                            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
-            
+
             for result in results:
                 conversations_list.append(
                     conversations + [{"role": "assistant", "content": result}])
-        else:            
+        else:
             for _ in range(self.args.human_model_num):
-                single_result = chat_with_continue(llm=self.llms[0], conversations=conversations, llm_config=llm_config)                
+                single_result = chat_with_continue(
+                    llm=self.llms[0], conversations=conversations, llm_config=llm_config)
                 results.append(single_result.content)
                 input_tokens_count += single_result.input_tokens_count
                 generated_tokens_count += single_result.generated_tokens_count
-                conversations_list.append(conversations + [{"role": "assistant", "content": single_result.content}])
-        
+                conversations_list.append(
+                    conversations + [{"role": "assistant", "content": single_result.content}])
+
         statistics = {
             "input_tokens_count": input_tokens_count,
             "generated_tokens_count": generated_tokens_count
-        }        
-    
+        }
+
         if self.args.request_id and not self.args.skip_events:
             _ = queue_communicate.send_event(
                 request_id=self.args.request_id,

autocoder/common/code_auto_generate_strict_diff.py

@@ -7,6 +7,8 @@ from autocoder.common import sys_prompt
 from concurrent.futures import ThreadPoolExecutor
 import json
 from autocoder.common.utils_code_auto_generate import chat_with_continue
+from autocoder.common.printer import Printer
+from autocoder.rag.token_counter import count_tokens
 
 class CodeAutoGenerateStrictDiff:
     def __init__(
@@ -309,6 +311,14 @@ class CodeAutoGenerateStrictDiff:
         results = []
         input_tokens_count = 0
         generated_tokens_count = 0
+
+        printer = Printer()
+        estimated_input_tokens = count_tokens(json.dumps(conversations, ensure_ascii=False))
+        printer.print_in_terminal("estimated_input_tokens_in_generate", style="yellow",
+                                  estimated_input_tokens_in_generate=estimated_input_tokens,
+                                  generate_mode="strict_diff"
+                                  )
+
         if not self.args.human_as_model:
             with ThreadPoolExecutor(max_workers=len(self.llms) * self.generate_times_same_model) as executor:
                 futures = []

autocoder/common/code_modification_ranker.py

@@ -8,6 +8,8 @@ from concurrent.futures import ThreadPoolExecutor, as_completed
 import traceback
 from autocoder.common.utils_code_auto_generate import chat_with_continue
 from byzerllm.utils.str2model import to_model
+
+from autocoder.utils.llms import get_llm_names
 class RankResult(BaseModel):
     rank_result: List[int]
 
@@ -78,9 +80,7 @@ class CodeModificationRanker:
                 # Submit tasks for each model and generate_times
                 futures = []
                 for llm in self.llms:                    
-                    model_name = getattr(llm, 'default_model_name', None)                    
-                    if not model_name:
-                        model_name = "unknown(without default model name)"
+                    model_name = ",".join(get_llm_names(llm))
                     self.printer.print_in_terminal(
                         "ranking_start", style="blue", count=len(generate_result.contents), model_name=model_name)
                     

autocoder/common/global_cancel.py

@@ -0,0 +1,21 @@
+import threading
+
+class GlobalCancel:
+    def __init__(self):
+        self._flag = False
+        self._lock = threading.Lock()
+    
+    @property
+    def requested(self):
+        with self._lock:
+            return self._flag
+    
+    def set(self):
+        with self._lock:
+            self._flag = True
+    
+    def reset(self):
+        with self._lock:
+            self._flag = False
+
+global_cancel = GlobalCancel()

autocoder/common/printer.py

@@ -32,7 +32,10 @@ class Printer:
             else:
                 self.console.print(format_str_jinja2(self.get_message_from_key(msg_key),**kwargs))
         except Exception as e:
-            print(self.get_message_from_key(msg_key))
+            try:
+                print(self.get_message_from_key(msg_key))
+            except Exception as e:
+                print(f"text key: {msg_key} not found")
 
     
     def print_str_in_terminal(self, content: str, style: str = None):     

autocoder/dispacher/actions/action.py

@@ -26,6 +26,7 @@ from autocoder.utils.conversation_store import store_code_model_conversation
 from loguru import logger
 import time
 from autocoder.common.printer import Printer
+from autocoder.utils.llms import get_llm_names
 
 
 class BaseAction:    
@@ -123,11 +124,16 @@ class ActionTSProject(BaseAction):
                 generate_result = generate.single_round_run(
                     query=args.query, source_content=content
                 )
+            elapsed_time = time.time() - start_time
+            speed = generate_result.metadata.get('generated_tokens_count', 0) / elapsed_time if elapsed_time > 0 else 0            
+            model_names = ",".join(get_llm_names(self.llm))
             self.printer.print_in_terminal(
                 "code_generation_complete",
-                duration=time.time() - start_time,
+                duration=elapsed_time,
                 input_tokens=generate_result.metadata.get('input_tokens_count', 0),
-                output_tokens=generate_result.metadata.get('generated_tokens_count', 0)
+                output_tokens=generate_result.metadata.get('generated_tokens_count', 0),
+                speed=round(speed, 2),
+                model_names=model_names
             )
             merge_result = None
             if args.execute and args.auto_merge:
@@ -213,11 +219,16 @@ class ActionPyScriptProject(BaseAction):
                     query=args.query, source_content=content
                 )
 
+            elapsed_time = time.time() - start_time
+            speed = generate_result.metadata.get('generated_tokens_count', 0) / elapsed_time if elapsed_time > 0 else 0
+            model_names = ",".join(get_llm_names(self.llm))
             self.printer.print_in_terminal(
                 "code_generation_complete",
-                duration=time.time() - start_time,
+                duration=elapsed_time,
                 input_tokens=generate_result.metadata.get('input_tokens_count', 0),
-                output_tokens=generate_result.metadata.get('generated_tokens_count', 0)
+                output_tokens=generate_result.metadata.get('generated_tokens_count', 0),
+                speed=round(speed, 2),
+                model_names=model_names
             )
             merge_result = None
             if args.execute and args.auto_merge:
@@ -335,11 +346,16 @@ class ActionPyProject(BaseAction):
                 generate_result = generate.single_round_run(
                     query=args.query, source_content=content
                 )
+            elapsed_time = time.time() - start_time
+            speed = generate_result.metadata.get('generated_tokens_count', 0) / elapsed_time if elapsed_time > 0 else 0
+            model_names = ",".join(get_llm_names(self.llm))
             self.printer.print_in_terminal(
                 "code_generation_complete",
-                duration=time.time() - start_time,
+                duration=elapsed_time,
                 input_tokens=generate_result.metadata.get('input_tokens_count', 0),
-                output_tokens=generate_result.metadata.get('generated_tokens_count', 0)
+                output_tokens=generate_result.metadata.get('generated_tokens_count', 0),
+                speed=round(speed, 2),
+                model_names=model_names
             )
             merge_result = None
             if args.execute and args.auto_merge:
@@ -440,11 +456,16 @@ class ActionSuffixProject(BaseAction):
                     query=args.query, source_content=content
                 )
               
+        elapsed_time = time.time() - start_time
+        speed = generate_result.metadata.get('generated_tokens_count', 0) / elapsed_time if elapsed_time > 0 else 0
+        model_names = ",".join(get_llm_names(self.llm))
         self.printer.print_in_terminal(
             "code_generation_complete",
-            duration=time.time() - start_time,
+            duration=elapsed_time,
             input_tokens=generate_result.metadata.get('input_tokens_count', 0),
-            output_tokens=generate_result.metadata.get('generated_tokens_count', 0)
+            output_tokens=generate_result.metadata.get('generated_tokens_count', 0),
+            speed=round(speed, 2),
+            model_names=model_names
         )
         merge_result = None
         if args.execute and args.auto_merge:

autocoder/dispacher/actions/plugins/action_regex_project.py

@@ -12,9 +12,10 @@ from autocoder.common.code_auto_generate_editblock import CodeAutoGenerateEditBl
 from autocoder.index.entry import build_index_and_filter_files
 from autocoder.regexproject import RegexProject
 from autocoder.utils.conversation_store import store_code_model_conversation
-from loguru import logger
+from autocoder.common.printer import Printer
 import time
-
+from autocoder.utils.llms import get_llm_names
+from loguru import logger
 class ActionRegexProject:
     def __init__(
         self, args: AutoCoderArgs, llm: Optional[byzerllm.ByzerLLM] = None
@@ -22,6 +23,7 @@ class ActionRegexProject:
         self.args = args
         self.llm = llm
         self.pp = None
+        self.printer = Printer()
 
     def run(self):
         args = self.args
@@ -58,7 +60,7 @@ class ActionRegexProject:
 
         start_time = time.time()
         if args.execute:
-            logger.info("Auto generate the code...")
+            self.printer.print_in_terminal("code_generation_start")
             
             if args.auto_merge == "diff":
                 generate = CodeAutoGenerateDiff(
@@ -83,10 +85,20 @@ class ActionRegexProject:
                     query=args.query, source_content=content
                 )
 
-            logger.info(f"Code generation completed in {time.time() - start_time:.2f} seconds, input_tokens_count: {generate_result.metadata.get('input_tokens_count', 0)}, generated_tokens_count: {generate_result.metadata.get('generated_tokens_count', 0)}")    
+            elapsed_time = time.time() - start_time
+            speed = generate_result.metadata.get('generated_tokens_count', 0) / elapsed_time if elapsed_time > 0 else 0
+            model_names = ",".join(get_llm_names(self.llm))
+            self.printer.print_in_terminal(
+                "code_generation_complete",
+                duration=elapsed_time,
+                input_tokens=generate_result.metadata.get('input_tokens_count', 0),
+                output_tokens=generate_result.metadata.get('generated_tokens_count', 0),
+                speed=round(speed, 2),
+                model_names=model_names
+            )
             merge_result = None
             if args.execute and args.auto_merge:
-                logger.info("Auto merge the code...")
+                self.printer.print_in_terminal("code_merge_start")
                 if args.auto_merge == "diff":
                     code_merge = CodeAutoMergeDiff(llm=self.llm, args=self.args)
                     merge_result = code_merge.merge_code(generate_result=generate_result)

autocoder/index/filter/quick_filter.py

@@ -17,6 +17,8 @@ from autocoder.common.printer import Printer
 from concurrent.futures import ThreadPoolExecutor
 import threading
 
+from autocoder.utils.llms import get_llm_names
+
 
 def get_file_path(file_path):
     if file_path.startswith("##"):
@@ -70,9 +72,7 @@ class QuickFilter():
 
         def process_chunk(chunk_index: int, chunk: List[IndexItem]) -> None:
             try:
-                model_name = getattr(self.index_manager.index_filter_llm, 'default_model_name', None)
-                if not model_name:
-                    model_name = "unknown(without default model name)"
+                model_name = ",".join(get_llm_names(self.index_manager.index_filter_llm))
                 
                 if chunk_index == 0:
                     # 第一个chunk使用流式输出
@@ -180,9 +180,7 @@ class QuickFilter():
             return self.big_filter(index_items)
         
         try:
-            model_name = getattr(self.index_manager.index_filter_llm, 'default_model_name', None)
-            if not model_name:
-                model_name = "unknown(without default model name)"
+            model_name = ",".join(get_llm_names(self.index_manager.index_filter_llm))
             
             # 渲染 Prompt 模板
             query = self.quick_filter_files.prompt(index_items, self.args.query)