Merge branch 'master' into frontier

2025-12-06 14:36:48 +00:00 · 2025-05-06 22:18:48 +08:00
--- a/config.py
+++ b/config.py
@@ -7,11 +7,11 @@
    Configuration reading priority: environment variable > config_private.py > config.py
 """

-# [step 1-1]>> ( 接入GPT等模型 ) API_KEY = "sk-123456789xxxxxxxxxxxxxxxxxxxxxxxxxxxxxx123456789"。极少数情况下，还需要填写组织（格式如org-123456789abcdefghijklmno的），请向下翻，找 API_ORG 设置项
+# [step 1-1]>> ( 接入OpenAI模型家族 ) API_KEY = "sk-123456789xxxxxxxxxxxxxxxxxxxxxxxxxxxxxx123456789"。极少数情况下，还需要填写组织（格式如org-123456789abcdefghijklmno的），请向下翻，找 API_ORG 设置项
 API_KEY = "在此处填写APIKEY"    # 可同时填写多个API-KEY，用英文逗号分割，例如API_KEY = "sk-openaikey1,sk-openaikey2,fkxxxx-api2dkey3,azure-apikey4"

-# [step 1-2]>> ( 接入通义 qwen-max ) 接入通义千问在线大模型，api-key获取地址 https://dashscope.console.aliyun.com/
-DASHSCOPE_API_KEY = "" # 阿里灵积云API_KEY
+# [step 1-2]>> ( 强烈推荐！接入通义家族 & 大模型服务平台百炼 ) 接入通义千问在线大模型，api-key获取地址 https://dashscope.console.aliyun.com/
+DASHSCOPE_API_KEY = "" # 阿里灵积云API_KEY（用于接入qwen-max，dashscope-qwen3-14b，dashscope-deepseek-r1等）

 # [step 1-3]>> ( 接入 deepseek-reasoner, 即 deepseek-r1 ) 深度求索(DeepSeek) API KEY，默认请求地址为"https://api.deepseek.com/v1/chat/completions"
 DEEPSEEK_API_KEY = ""
@@ -46,6 +46,7 @@ AVAIL_LLM_MODELS = ["qwen-max", "o1-mini", "o1-mini-2024-09-12", "o1", "o1-2024-
                    "deepseek-chat", "deepseek-coder", "deepseek-reasoner", 
                    "volcengine-deepseek-r1-250120", "volcengine-deepseek-v3-241226",
                    "dashscope-deepseek-r1", "dashscope-deepseek-v3",
+                    "dashscope-qwen3-14b", "dashscope-qwen3-235b-a22b", "dashscope-qwen3-32b",
                    ]

 EMBEDDING_MODEL = "text-embedding-3-small"
--- a/main.py
+++ b/main.py
@@ -34,7 +34,7 @@ def encode_plugin_info(k, plugin)->str:

 def main():
    import gradio as gr
-    if gr.__version__ not in ['3.32.12', '3.32.13']:
+    if gr.__version__ not in ['3.32.14', '3.32.13']:
        raise ModuleNotFoundError("使用项目内置Gradio获取最优体验! 请运行 `pip install -r requirements.txt` 指令安装内置Gradio及其他依赖, 详情信息见requirements.txt.")

    # 一些基础工具
--- a/request_llms/bridge_all.py
+++ b/request_llms/bridge_all.py
@@ -282,6 +282,7 @@ model_info = {
        "fn_with_ui": chatgpt_ui,
        "fn_without_ui": chatgpt_noui,
        "endpoint": openai_endpoint,
+        "can_multi_thread": True,
        "max_token": 128000,
        "tokenizer": tokenizer_gpt4,
        "token_cnt": get_token_num_gpt4,
@@ -314,6 +315,50 @@ model_info = {
        "openai_force_temperature_one": True,
    },

+    "gpt-4.1":{
+        "fn_with_ui": chatgpt_ui,
+        "fn_without_ui": chatgpt_noui,
+        "has_multimodal_capacity": True,
+        "endpoint": openai_endpoint,
+        "max_token": 828000,
+        "tokenizer": tokenizer_gpt4,
+        "token_cnt": get_token_num_gpt4,
+    },
+
+    "gpt-4.1-mini":{
+        "fn_with_ui": chatgpt_ui,
+        "fn_without_ui": chatgpt_noui,
+        "has_multimodal_capacity": True,
+        "endpoint": openai_endpoint,
+        "max_token": 828000,
+        "tokenizer": tokenizer_gpt4,
+        "token_cnt": get_token_num_gpt4,
+    },
+
+    "o3":{
+        "fn_with_ui": chatgpt_ui,
+        "fn_without_ui": chatgpt_noui,
+        "has_multimodal_capacity": True,
+        "endpoint": openai_endpoint,
+        "max_token": 828000,
+        "tokenizer": tokenizer_gpt4,
+        "token_cnt": get_token_num_gpt4,
+        "openai_disable_system_prompt": True,
+        "openai_disable_stream": True,
+        "openai_force_temperature_one": True,
+    },
+
+    "o4-mini":{
+        "fn_with_ui": chatgpt_ui,
+        "fn_without_ui": chatgpt_noui,
+        "has_multimodal_capacity": True,
+        "can_multi_thread": True,
+        "endpoint": openai_endpoint,
+        "max_token": 828000,
+        "tokenizer": tokenizer_gpt4,
+        "token_cnt": get_token_num_gpt4,
+    },
+
    "gpt-4-turbo": {
        "fn_with_ui": chatgpt_ui,
        "fn_without_ui": chatgpt_noui,
@@ -824,7 +869,10 @@ if "qwen-local" in AVAIL_LLM_MODELS:
        logger.error(trimmed_format_exc())

 # -=-=-=-=-=-=- 阿里云百炼（通义）-在线模型 -=-=-=-=-=-=-
-qwen_models = ["qwen-max-latest", "qwen-max-2025-01-25","qwen-max","qwen-turbo","qwen-plus","dashscope-deepseek-r1","dashscope-deepseek-v3"]
+qwen_models = ["qwen-max-latest", "qwen-max-2025-01-25","qwen-max","qwen-turbo","qwen-plus",
+               "dashscope-deepseek-r1","dashscope-deepseek-v3",
+               "dashscope-qwen3-14b", "dashscope-qwen3-235b-a22b", "dashscope-qwen3-qwen3-32b",
+               ]
 if any(item in qwen_models for item in AVAIL_LLM_MODELS):
    try:
        from .bridge_qwen import predict_no_ui_long_connection as qwen_noui
@@ -893,6 +941,34 @@ if any(item in qwen_models for item in AVAIL_LLM_MODELS):
                "max_token": 57344,
                "tokenizer": tokenizer_gpt35,
                "token_cnt": get_token_num_gpt35,
+            },
+            "dashscope-qwen3-14b": {
+                "fn_with_ui": qwen_ui,
+                "fn_without_ui": qwen_noui,
+                "enable_reasoning": True,
+                "can_multi_thread": True,
+                "endpoint": None,
+                "max_token": 129024,
+                "tokenizer": tokenizer_gpt35,
+                "token_cnt": get_token_num_gpt35,
+            },
+            "dashscope-qwen3-235b-a22b": {
+                "fn_with_ui": qwen_ui,
+                "fn_without_ui": qwen_noui,
+                "can_multi_thread": True,
+                "endpoint": None,
+                "max_token": 129024,
+                "tokenizer": tokenizer_gpt35,
+                "token_cnt": get_token_num_gpt35,
+            },
+            "dashscope-qwen3-32b": {
+                "fn_with_ui": qwen_ui,
+                "fn_without_ui": qwen_noui,
+                "can_multi_thread": True,
+                "endpoint": None,
+                "max_token": 129024,
+                "tokenizer": tokenizer_gpt35,
+                "token_cnt": get_token_num_gpt35,
            }
        })
    except:
--- a/requirements.txt
+++ b/requirements.txt
@@ -1,4 +1,4 @@
-https://public.agent-matrix.com/publish/gradio-3.32.13-py3-none-any.whl
+https://public.agent-matrix.com/publish/gradio-3.32.14-py3-none-any.whl
 fastapi==0.110
 gradio-client==0.8
 pypdf2==2.12.1
--- a/shared_utils/key_pattern_manager.py
+++ b/shared_utils/key_pattern_manager.py
@@ -82,6 +82,14 @@ def what_keys(keys):

    return f"检测到： OpenAI Key {avail_key_list['OpenAI Key']} 个, Azure Key {avail_key_list['Azure Key']} 个, API2D Key {avail_key_list['API2D Key']} 个"

+def is_o_family_for_openai(llm_model):
+    if not llm_model.startswith('o'):
+        return False
+    if llm_model in ['o1', 'o2', 'o3', 'o4', 'o5', 'o6', 'o7', 'o8']:
+        return True
+    if llm_model[:3] in ['o1-', 'o2-', 'o3-', 'o4-', 'o5-', 'o6-', 'o7-', 'o8-']:
+        return True
+    return False

 def select_api_key(keys, llm_model):
    import random
@@ -89,7 +97,7 @@ def select_api_key(keys, llm_model):
    key_list = keys.split(',')

    if llm_model.startswith('gpt-') or llm_model.startswith('chatgpt-') or \
-       llm_model.startswith('one-api-') or llm_model == 'o1' or llm_model.startswith('o1-'):
+       llm_model.startswith('one-api-') or is_o_family_for_openai(llm_model):
        for k in key_list:
            if is_openai_api_key(k): avail_key_list.append(k)

--- a/themes/common.js
+++ b/themes/common.js
@@ -1184,6 +1184,15 @@ async function clear_conversation(a, b, c) {
    update_conversation_metadata();
    let stopButton = document.getElementById("elem_stop");
    stopButton.click();
+    // Save back to local storage
+    try {
+        const EVENT_NAME = "gptac_reset_btn_clicked";
+        window.dispatchEvent(
+            new CustomEvent(EVENT_NAME, {
+                detail: ""
+            })
+        );
+    } catch (e) {}
    return reset_conversation(a, b);
 }