chunzhi/chatgpt_academic
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity

merge frontier branch (#1620)

Browse Source 
* Zhipu sdk update 适配最新的智谱SDK，支持GLM4v (#1502)

* 适配 google gemini 优化为从用户input中提取文件

* 适配最新的智谱SDK、支持glm-4v

* requirements.txt fix

* pending history check

---------

Co-authored-by: binary-husky <qingxu.fu@outlook.com>

* Update "生成多种Mermaid图表" plugin: Separate out the file reading function (#1520)

* Update crazy_functional.py with new functionality deal with PDF

* Update crazy_functional.py and Mermaid.py for plugin_kwargs

* Update crazy_functional.py with new chart type: mind map

* Update SELECT_PROMPT and i_say_show_user messages

* Update ArgsReminder message in get_crazy_functions() function

* Update with read md file and update PROMPTS

* Return the PROMPTS as the test found that the initial version worked best

* Update Mermaid chart generation function

* version 3.71

* 解决issues #1510

* Remove unnecessary text from sys_prompt in 解析历史输入 function

* Remove sys_prompt message in 解析历史输入 function

* Update bridge_all.py: supports gpt-4-turbo-preview (#1517)

* Update bridge_all.py: supports gpt-4-turbo-preview

supports gpt-4-turbo-preview

* Update bridge_all.py

---------

Co-authored-by: binary-husky <96192199+binary-husky@users.noreply.github.com>

* Update config.py: supports gpt-4-turbo-preview (#1516)

* Update config.py: supports gpt-4-turbo-preview

supports gpt-4-turbo-preview

* Update config.py

---------

Co-authored-by: binary-husky <96192199+binary-husky@users.noreply.github.com>

* Refactor 解析历史输入 function to handle file input

* Update Mermaid chart generation functionality

* rename files and functions

---------

Co-authored-by: binary-husky <qingxu.fu@outlook.com>
Co-authored-by: hongyi-zhao <hongyi.zhao@gmail.com>
Co-authored-by: binary-husky <96192199+binary-husky@users.noreply.github.com>

* 接入mathpix ocr功能 (#1468)

* Update Latex输出PDF结果.py

借助mathpix实现了PDF翻译中文并重新编译PDF

* Update config.py

add mathpix appid & appkey

* Add 'PDF翻译中文并重新编译PDF' feature to plugins.

---------

Co-authored-by: binary-husky <96192199+binary-husky@users.noreply.github.com>

* fix zhipuai

* check picture

* remove glm-4 due to bug

* 修改config

* 检查MATHPIX_APPID

* Remove unnecessary code and update
function_plugins dictionary

* capture non-standard token overflow

* bug fix #1524

* change mermaid style

* 支持mermaid 滚动放大缩小重置,鼠标滚动和拖拽 (#1530)

* 支持mermaid 滚动放大缩小重置,鼠标滚动和拖拽

* 微调未果 先stage一下

* update

---------

Co-authored-by: binary-husky <qingxu.fu@outlook.com>
Co-authored-by: binary-husky <96192199+binary-husky@users.noreply.github.com>

* ver 3.72

* change live2d

* save the status of ``clear btn` in cookie

* 前端选择保持

* js ui bug fix

* reset btn bug fix

* update live2d tips

* fix missing get_token_num method

* fix live2d toggle switch

* fix persistent custom btn with cookie

* fix zhipuai feedback with core functionality

* Refactor button update and clean up functions

* tailing space removal

* Fix missing MATHPIX_APPID and MATHPIX_APPKEY
configuration

* Prompt fix、脑图提示词优化 (#1537)

* 适配 google gemini 优化为从用户input中提取文件

* 脑图提示词优化

* Fix missing MATHPIX_APPID and MATHPIX_APPKEY
configuration

---------

Co-authored-by: binary-husky <qingxu.fu@outlook.com>

* 优化“PDF翻译中文并重新编译PDF”插件 (#1602)

* Add gemini_endpoint to API_URL_REDIRECT (#1560)

* Add gemini_endpoint to API_URL_REDIRECT

* Update gemini-pro and gemini-pro-vision model_info
endpoints

* Update to support new claude models (#1606)

* Add anthropic library and update claude models

* 更新bridge_claude.py文件，添加了对图片输入的支持。修复了一些bug。

* 添加Claude_3_Models变量以限制图片数量

* Refactor code to improve readability and
maintainability

* minor claude bug fix

* more flexible one-api support

* reformat config

* fix one-api new access bug

* dummy

* compat non-standard api

* version 3.73

---------

Co-authored-by: XIao <46100050+Kilig947@users.noreply.github.com>
Co-authored-by: Menghuan1918 <menghuan2003@outlook.com>
Co-authored-by: hongyi-zhao <hongyi.zhao@gmail.com>
Co-authored-by: Hao Ma <893017927@qq.com>
Co-authored-by: zeyuan huang <599012428@qq.com>
This commit is contained in:
binary-husky 2024-03-11 17:26:09 +08:00 committed by GitHub
parent cd18663800
commit c3140ce344
No known key found for this signature in database
GPG Key ID: B5690EEEBB952194
85 changed files with 866 additions and 642 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

4
check_proxy.py
View File

@ -47,7 +47,7 @@ def backup_and_download(current_version, remote_version):
shutil.copytree('./', backup_dir, ignore=lambda x, y: ['history']) shutil.copytree('./', backup_dir, ignore=lambda x, y: ['history'])
proxies = get_conf('proxies') proxies = get_conf('proxies')
try: r = requests.get('https://github.com/binary-husky/chatgpt_academic/archive/refs/heads/master.zip', proxies=proxies, stream=True) try: r = requests.get('https://github.com/binary-husky/chatgpt_academic/archive/refs/heads/master.zip', proxies=proxies, stream=True)
except: r = requests.get('https://public.gpt-academic.top/publish/master.zip', proxies=proxies, stream=True) except: r = requests.get('https://public.agent-matrix.com/publish/master.zip', proxies=proxies, stream=True)
zip_file_path = backup_dir+'/master.zip' zip_file_path = backup_dir+'/master.zip'
with open(zip_file_path, 'wb+') as f: with open(zip_file_path, 'wb+') as f:
f.write(r.content) f.write(r.content)
@ -113,7 +113,7 @@ def auto_update(raise_error=False):
import json import json
proxies = get_conf('proxies') proxies = get_conf('proxies')
try: response = requests.get("https://raw.githubusercontent.com/binary-husky/chatgpt_academic/master/version", proxies=proxies, timeout=5) try: response = requests.get("https://raw.githubusercontent.com/binary-husky/chatgpt_academic/master/version", proxies=proxies, timeout=5)
except: response = requests.get("https://public.gpt-academic.top/publish/version", proxies=proxies, timeout=5) except: response = requests.get("https://public.agent-matrix.com/publish/version", proxies=proxies, timeout=5)
remote_json_data = json.loads(response.text) remote_json_data = json.loads(response.text)
remote_version = remote_json_data['version'] remote_version = remote_json_data['version']
if remote_json_data["show_feature"]: if remote_json_data["show_feature"]:

75
config.py
View File

@ -30,7 +30,32 @@ if USE_PROXY:
else: else:
proxies = None proxies = None
# ------------------------------------ 以下配置可以优化体验, 但大部分场合下并不需要修改 ------------------------------------ # [step 3]>> 模型选择是 (注意: LLM_MODEL是默认选中的模型, 它*必须*被包含在AVAIL_LLM_MODELS列表中 )
LLM_MODEL = "gpt-3.5-turbo-16k" # 可选 ↓↓↓
AVAIL_LLM_MODELS = ["gpt-4-1106-preview", "gpt-4-turbo-preview", "gpt-4-vision-preview",
"gpt-3.5-turbo-1106", "gpt-3.5-turbo-16k", "gpt-3.5-turbo", "azure-gpt-3.5",
"gpt-4", "gpt-4-32k", "azure-gpt-4", "glm-4", "glm-3-turbo",
"gemini-pro", "chatglm3"
]
# --- --- --- ---
# P.S. 其他可用的模型还包括
# AVAIL_LLM_MODELS = [
# "qianfan", "deepseekcoder",
# "spark", "sparkv2", "sparkv3", "sparkv3.5",
# "qwen-turbo", "qwen-plus", "qwen-max", "qwen-local",
# "moonshot-v1-128k", "moonshot-v1-32k", "moonshot-v1-8k",
# "gpt-3.5-turbo-0613", "gpt-3.5-turbo-16k-0613", "gpt-3.5-turbo-0125"
# "claude-3-sonnet-20240229","claude-3-opus-20240229", "claude-2.1", "claude-instant-1.2",
# "moss", "llama2", "chatglm_onnx", "internlm", "jittorllms_pangualpha", "jittorllms_llama",
# ]
# --- --- --- ---
# 此外，为了更灵活地接入one-api多模型管理界面，您还可以在接入one-api时，
# 使用"one-api-*"前缀直接使用非标准方式接入的模型，例如
# AVAIL_LLM_MODELS = ["one-api-claude-3-sonnet-20240229(max_token=100000)"]
# --- --- --- ---
# --------------- 以下配置可以优化体验 ---------------
# 重新URL重新定向，实现更换API_URL的作用（高危设置! 常规情况下不要修改! 通过修改此设置，您将把您的API-KEY和对话隐私完全暴露给您设定的中间人！） # 重新URL重新定向，实现更换API_URL的作用（高危设置! 常规情况下不要修改! 通过修改此设置，您将把您的API-KEY和对话隐私完全暴露给您设定的中间人！）
# 格式: API_URL_REDIRECT = {"https://api.openai.com/v1/chat/completions": "在这里填写重定向的api.openai.com的URL"} # 格式: API_URL_REDIRECT = {"https://api.openai.com/v1/chat/completions": "在这里填写重定向的api.openai.com的URL"}
@ -85,22 +110,6 @@ MAX_RETRY = 2
DEFAULT_FN_GROUPS = ['对话', '编程', '学术', '智能体'] DEFAULT_FN_GROUPS = ['对话', '编程', '学术', '智能体']
# 模型选择是 (注意: LLM_MODEL是默认选中的模型, 它*必须*被包含在AVAIL_LLM_MODELS列表中 )
LLM_MODEL = "gpt-3.5-turbo-16k" # 可选 ↓↓↓
AVAIL_LLM_MODELS = ["gpt-4-1106-preview", "gpt-4-turbo-preview", "gpt-4-vision-preview",
"gpt-3.5-turbo-1106", "gpt-3.5-turbo-16k", "gpt-3.5-turbo", "azure-gpt-3.5",
"gpt-4", "gpt-4-32k", "azure-gpt-4", "glm-4", "glm-3-turbo",
"gemini-pro", "chatglm3", "claude-2"]
# P.S. 其他可用的模型还包括 [
# "moonshot-v1-128k", "moonshot-v1-32k", "moonshot-v1-8k",
# "qwen-turbo", "qwen-plus", "qwen-max",
# "zhipuai", "qianfan", "deepseekcoder", "llama2", "qwen-local", "gpt-3.5-turbo-0613", "moss",
# "gpt-3.5-turbo-16k-0613", "gpt-3.5-random", "api2d-gpt-3.5-turbo", 'api2d-gpt-3.5-turbo-16k',
# "spark", "sparkv2", "sparkv3", "sparkv3.5",
# "chatglm_onnx", "claude-1-100k", "claude-2", "internlm", "jittorllms_pangualpha", "jittorllms_llama"
# ]
# 定义界面上“询问多个GPT模型”插件应该使用哪些模型，请从AVAIL_LLM_MODELS中选择，并在不同模型之间用`&`间隔，例如"gpt-3.5-turbo&chatglm3&azure-gpt-4" # 定义界面上“询问多个GPT模型”插件应该使用哪些模型，请从AVAIL_LLM_MODELS中选择，并在不同模型之间用`&`间隔，例如"gpt-3.5-turbo&chatglm3&azure-gpt-4"
MULTI_QUERY_LLM_MODELS = "gpt-3.5-turbo&chatglm3" MULTI_QUERY_LLM_MODELS = "gpt-3.5-turbo&chatglm3"
@ -129,6 +138,7 @@ CHATGLM_PTUNING_CHECKPOINT = "" # 例如"/home/hmp/ChatGLM2-6B/ptuning/output/6b
LOCAL_MODEL_DEVICE = "cpu" # 可选 "cuda" LOCAL_MODEL_DEVICE = "cpu" # 可选 "cuda"
LOCAL_MODEL_QUANT = "FP16" # 默认 "FP16" "INT4" 启用量化INT4版本 "INT8" 启用量化INT8版本 LOCAL_MODEL_QUANT = "FP16" # 默认 "FP16" "INT4" 启用量化INT4版本 "INT8" 启用量化INT8版本
# 设置gradio的并行线程数（不需要修改） # 设置gradio的并行线程数（不需要修改）
CONCURRENT_COUNT = 100 CONCURRENT_COUNT = 100
@ -174,14 +184,8 @@ AZURE_ENGINE = "填入你亲手写的部署名" # 读 docs\use_azure.
AZURE_CFG_ARRAY = {} AZURE_CFG_ARRAY = {}
# 使用Newbing (不推荐使用，未来将删除) # 阿里云实时语音识别 配置难度较高
NEWBING_STYLE = "creative" # ["creative", "balanced", "precise"] # 参考 https://github.com/binary-husky/gpt_academic/blob/master/docs/use_audio.md
NEWBING_COOKIES = """
put your new bing cookies here
"""
# 阿里云实时语音识别 配置难度较高 仅建议高手用户使用 参考 https://github.com/binary-husky/gpt_academic/blob/master/docs/use_audio.md
ENABLE_AUDIO = False ENABLE_AUDIO = False
ALIYUN_TOKEN="" # 例如 f37f30e0f9934c34a992f6f64f7eba4f ALIYUN_TOKEN="" # 例如 f37f30e0f9934c34a992f6f64f7eba4f
ALIYUN_APPKEY="" # 例如 RoPlZrM88DnAFkZK ALIYUN_APPKEY="" # 例如 RoPlZrM88DnAFkZK
@ -200,18 +204,14 @@ ZHIPUAI_API_KEY = ""
ZHIPUAI_MODEL = "" # 此选项已废弃，不再需要填写 ZHIPUAI_MODEL = "" # 此选项已废弃，不再需要填写
# # 火山引擎YUNQUE大模型
# YUNQUE_SECRET_KEY = ""
# YUNQUE_ACCESS_KEY = ""
# YUNQUE_MODEL = ""
# Claude API KEY # Claude API KEY
ANTHROPIC_API_KEY = "" ANTHROPIC_API_KEY = ""
# 月之暗面 API KEY # 月之暗面 API KEY
MOONSHOT_API_KEY = "" MOONSHOT_API_KEY = ""
# Mathpix 拥有执行PDF的OCR功能，但是需要注册账号 # Mathpix 拥有执行PDF的OCR功能，但是需要注册账号
MATHPIX_APPID = "" MATHPIX_APPID = ""
MATHPIX_APPKEY = "" MATHPIX_APPKEY = ""
@ -270,7 +270,11 @@ PLUGIN_HOT_RELOAD = False
# 自定义按钮的最大数量限制 # 自定义按钮的最大数量限制
NUM_CUSTOM_BASIC_BTN = 4 NUM_CUSTOM_BASIC_BTN = 4
""" """
--------------- 配置关联关系说明 ---------------
在线大模型配置关联关系示意图 在线大模型配置关联关系示意图
│ │
├── "gpt-3.5-turbo" 等openai模型 ├── "gpt-3.5-turbo" 等openai模型
@ -294,7 +298,7 @@ NUM_CUSTOM_BASIC_BTN = 4
│ ├── XFYUN_API_SECRET │ ├── XFYUN_API_SECRET
│ └── XFYUN_API_KEY │ └── XFYUN_API_KEY
│ │
├── "claude-1-100k" 等claude模型 ├── "claude-3-opus-20240229" 等claude模型
│ └── ANTHROPIC_API_KEY │ └── ANTHROPIC_API_KEY
│ │
├── "stack-claude" ├── "stack-claude"
@ -315,9 +319,10 @@ NUM_CUSTOM_BASIC_BTN = 4
├── "Gemini" ├── "Gemini"
│ └── GEMINI_API_KEY │ └── GEMINI_API_KEY
│ │
└── "newbing" Newbing接口不再稳定，不推荐使用 └── "one-api-...(max_token=...)" 用一种更方便的方式接入one-api多模型管理界面
├── NEWBING_STYLE ├── AVAIL_LLM_MODELS
└── NEWBING_COOKIES ├── API_KEY
└── API_URL_REDIRECT
本地大模型示意图 本地大模型示意图

8
core_functional.py
View File

@ -38,12 +38,12 @@ def get_core_functions():
"总结绘制脑图": { "总结绘制脑图": {
# 前缀，会被加在你的输入之前。例如，用来描述你的要求，例如翻译、解释代码、润色等等 # 前缀，会被加在你的输入之前。例如，用来描述你的要求，例如翻译、解释代码、润色等等
"Prefix": r"", "Prefix": '''"""\n\n''',
# 后缀，会被加在你的输入之后。例如，配合前缀可以把你的输入内容用引号圈起来 # 后缀，会被加在你的输入之后。例如，配合前缀可以把你的输入内容用引号圈起来
"Suffix": "Suffix":
# dedent() 函数用于去除多行字符串的缩进 # dedent() 函数用于去除多行字符串的缩进
dedent("\n"+r''' dedent("\n\n"+r'''
============================== """
使用mermaid flowchart对以上文本进行总结，概括上述段落的内容以及内在逻辑关系，例如： 使用mermaid flowchart对以上文本进行总结，概括上述段落的内容以及内在逻辑关系，例如：
@ -57,7 +57,7 @@ def get_core_functions():
C --> |"箭头名2"| F["节点名6"] C --> |"箭头名2"| F["节点名6"]
``` ```
警告： 注意：
（1）使用中文 （1）使用中文
（2）节点名字使用引号包裹，如["Laptop"] （2）节点名字使用引号包裹，如["Laptop"]
（3）`|` 和 `"`之间不要存在空格 （3）`|` 和 `"`之间不要存在空格

130
crazy_functions/Latex输出PDF.py
View File

@ -1,4 +1,4 @@
from toolbox import update_ui, trimmed_format_exc, get_conf, get_log_folder, promote_file_to_downloadzone from toolbox import update_ui, trimmed_format_exc, get_conf, get_log_folder, promote_file_to_downloadzone, check_repeat_upload, map_file_to_sha256
from toolbox import CatchException, report_exception, update_ui_lastest_msg, zip_result, gen_time_str from toolbox import CatchException, report_exception, update_ui_lastest_msg, zip_result, gen_time_str
from functools import partial from functools import partial
import glob, os, requests, time, json, tarfile import glob, os, requests, time, json, tarfile
@ -438,47 +438,101 @@ def PDF翻译中文并重新编译PDF(txt, llm_kwargs, plugin_kwargs, chatbot, h
yield from update_ui(chatbot=chatbot, history=history) # 刷新界面 yield from update_ui(chatbot=chatbot, history=history) # 刷新界面
return return
# <-------------- convert pdf into tex -------------> hash_tag = map_file_to_sha256(file_manifest[0])
project_folder = pdf2tex_project(file_manifest[0])
# Translate English Latex to Chinese Latex, and compile it # <-------------- check repeated pdf ------------->
file_manifest = [f for f in glob.glob(f'{project_folder}/**/*.tex', recursive=True)] chatbot.append([f"检查PDF是否被重复上传", "正在检查..."])
if len(file_manifest) == 0: yield from update_ui(chatbot=chatbot, history=history)
report_exception(chatbot, history, a=f"解析项目: {txt}", b=f"找不到任何.tex文件: {txt}") repeat, project_folder = check_repeat_upload(file_manifest[0], hash_tag)
yield from update_ui(chatbot=chatbot, history=history) # 刷新界面
return
# <-------------- if is a zip/tar file -------------> except_flag = False
project_folder = desend_to_extracted_folder_if_exist(project_folder)
# <-------------- move latex project away from temp folder -------------> if repeat:
project_folder = move_project(project_folder) yield from update_ui_lastest_msg(f"发现重复上传，请查收结果（压缩包）...", chatbot=chatbot, history=history)
# <-------------- if merge_translate_zh is already generated, skip gpt req -------------> try:
if not os.path.exists(project_folder + '/merge_translate_zh.tex'): trans_html_file = [f for f in glob.glob(f'{project_folder}/**/*.trans.html', recursive=True)][0]
yield from Latex精细分解与转化(file_manifest, project_folder, llm_kwargs, plugin_kwargs, promote_file_to_downloadzone(trans_html_file, rename_file=None, chatbot=chatbot)
chatbot, history, system_prompt, mode='translate_zh',
switch_prompt=_switch_prompt_)
# <-------------- compile PDF -------------> translate_pdf = [f for f in glob.glob(f'{project_folder}/**/merge_translate_zh.pdf', recursive=True)][0]
success = yield from 编译Latex(chatbot, history, main_file_original='merge', promote_file_to_downloadzone(translate_pdf, rename_file=None, chatbot=chatbot)
main_file_modified='merge_translate_zh', mode='translate_zh',
work_folder_original=project_folder, work_folder_modified=project_folder,
work_folder=project_folder)
# <-------------- zip PDF -------------> comparison_pdf = [f for f in glob.glob(f'{project_folder}/**/comparison.pdf', recursive=True)][0]
zip_res = zip_result(project_folder) promote_file_to_downloadzone(comparison_pdf, rename_file=None, chatbot=chatbot)
if success:
chatbot.append((f"成功啦", '请查收结果（压缩包）...'))
yield from update_ui(chatbot=chatbot, history=history);
time.sleep(1) # 刷新界面
promote_file_to_downloadzone(file=zip_res, chatbot=chatbot)
else:
chatbot.append((f"失败了",
'虽然PDF生成失败了, 但请查收结果（压缩包）, 内含已经翻译的Tex文档, 您可以到Github Issue区, 用该压缩包进行反馈。如系统是Linux，请检查系统字体（见Github wiki） ...'))
yield from update_ui(chatbot=chatbot, history=history);
time.sleep(1) # 刷新界面
promote_file_to_downloadzone(file=zip_res, chatbot=chatbot)
# <-------------- we are done -------------> zip_res = zip_result(project_folder)
return success promote_file_to_downloadzone(file=zip_res, chatbot=chatbot)
return True
except:
report_exception(chatbot, history, b=f"发现重复上传，但是无法找到相关文件")
yield from update_ui(chatbot=chatbot, history=history)
chatbot.append([f"没有相关文件", '尝试重新翻译PDF...'])
yield from update_ui(chatbot=chatbot, history=history)
except_flag = True
elif not repeat or except_flag:
yield from update_ui_lastest_msg(f"未发现重复上传", chatbot=chatbot, history=history)
# <-------------- convert pdf into tex ------------->
chatbot.append([f"解析项目: {txt}", "正在将PDF转换为tex项目，请耐心等待..."])
yield from update_ui(chatbot=chatbot, history=history)
project_folder = pdf2tex_project(file_manifest[0])
if project_folder is None:
report_exception(chatbot, history, a=f"解析项目: {txt}", b=f"PDF转换为tex项目失败")
yield from update_ui(chatbot=chatbot, history=history)
return False
# <-------------- translate latex file into Chinese ------------->
yield from update_ui_lastest_msg("正在tex项目将翻译为中文...", chatbot=chatbot, history=history)
file_manifest = [f for f in glob.glob(f'{project_folder}/**/*.tex', recursive=True)]
if len(file_manifest) == 0:
report_exception(chatbot, history, a=f"解析项目: {txt}", b=f"找不到任何.tex文件: {txt}")
yield from update_ui(chatbot=chatbot, history=history) # 刷新界面
return
# <-------------- if is a zip/tar file ------------->
project_folder = desend_to_extracted_folder_if_exist(project_folder)
# <-------------- move latex project away from temp folder ------------->
project_folder = move_project(project_folder)
# <-------------- set a hash tag for repeat-checking ------------->
with open(pj(project_folder, hash_tag + '.tag'), 'w') as f:
f.write(hash_tag)
f.close()
# <-------------- if merge_translate_zh is already generated, skip gpt req ------------->
if not os.path.exists(project_folder + '/merge_translate_zh.tex'):
yield from Latex精细分解与转化(file_manifest, project_folder, llm_kwargs, plugin_kwargs,
chatbot, history, system_prompt, mode='translate_zh',
switch_prompt=_switch_prompt_)
# <-------------- compile PDF ------------->
yield from update_ui_lastest_msg("正在将翻译好的项目tex项目编译为PDF...", chatbot=chatbot, history=history)
success = yield from 编译Latex(chatbot, history, main_file_original='merge',
main_file_modified='merge_translate_zh', mode='translate_zh',
work_folder_original=project_folder, work_folder_modified=project_folder,
work_folder=project_folder)
# <-------------- zip PDF ------------->
zip_res = zip_result(project_folder)
if success:
chatbot.append((f"成功啦", '请查收结果（压缩包）...'))
yield from update_ui(chatbot=chatbot, history=history);
time.sleep(1) # 刷新界面
promote_file_to_downloadzone(file=zip_res, chatbot=chatbot)
else:
chatbot.append((f"失败了",
'虽然PDF生成失败了, 但请查收结果（压缩包）, 内含已经翻译的Tex文档, 您可以到Github Issue区, 用该压缩包进行反馈。如系统是Linux，请检查系统字体（见Github wiki） ...'))
yield from update_ui(chatbot=chatbot, history=history);
time.sleep(1) # 刷新界面
promote_file_to_downloadzone(file=zip_res, chatbot=chatbot)
# <-------------- we are done ------------->
return success

98
request_llms/bridge_all.py
View File

@ -8,10 +8,10 @@
具备多线程调用能力的函数：在函数插件中被调用，灵活而简洁 具备多线程调用能力的函数：在函数插件中被调用，灵活而简洁
2. predict_no_ui_long_connection(...) 2. predict_no_ui_long_connection(...)
""" """
import tiktoken, copy import tiktoken, copy, re
from functools import lru_cache from functools import lru_cache
from concurrent.futures import ThreadPoolExecutor from concurrent.futures import ThreadPoolExecutor
from toolbox import get_conf, trimmed_format_exc, apply_gpt_academic_string_mask from toolbox import get_conf, trimmed_format_exc, apply_gpt_academic_string_mask, read_one_api_model_name
from .bridge_chatgpt import predict_no_ui_long_connection as chatgpt_noui from .bridge_chatgpt import predict_no_ui_long_connection as chatgpt_noui
from .bridge_chatgpt import predict as chatgpt_ui from .bridge_chatgpt import predict as chatgpt_ui
@ -61,6 +61,9 @@ API_URL_REDIRECT, AZURE_ENDPOINT, AZURE_ENGINE = get_conf("API_URL_REDIRECT", "A
openai_endpoint = "https://api.openai.com/v1/chat/completions" openai_endpoint = "https://api.openai.com/v1/chat/completions"
api2d_endpoint = "https://openai.api2d.net/v1/chat/completions" api2d_endpoint = "https://openai.api2d.net/v1/chat/completions"
newbing_endpoint = "wss://sydney.bing.com/sydney/ChatHub" newbing_endpoint = "wss://sydney.bing.com/sydney/ChatHub"
gemini_endpoint = "https://generativelanguage.googleapis.com/v1beta/models"
claude_endpoint = "https://api.anthropic.com"
if not AZURE_ENDPOINT.endswith('/'): AZURE_ENDPOINT += '/' if not AZURE_ENDPOINT.endswith('/'): AZURE_ENDPOINT += '/'
azure_endpoint = AZURE_ENDPOINT + f'openai/deployments/{AZURE_ENGINE}/chat/completions?api-version=2023-05-15' azure_endpoint = AZURE_ENDPOINT + f'openai/deployments/{AZURE_ENGINE}/chat/completions?api-version=2023-05-15'
# 兼容旧版的配置 # 兼容旧版的配置
@ -75,7 +78,8 @@ except:
if openai_endpoint in API_URL_REDIRECT: openai_endpoint = API_URL_REDIRECT[openai_endpoint] if openai_endpoint in API_URL_REDIRECT: openai_endpoint = API_URL_REDIRECT[openai_endpoint]
if api2d_endpoint in API_URL_REDIRECT: api2d_endpoint = API_URL_REDIRECT[api2d_endpoint] if api2d_endpoint in API_URL_REDIRECT: api2d_endpoint = API_URL_REDIRECT[api2d_endpoint]
if newbing_endpoint in API_URL_REDIRECT: newbing_endpoint = API_URL_REDIRECT[newbing_endpoint] if newbing_endpoint in API_URL_REDIRECT: newbing_endpoint = API_URL_REDIRECT[newbing_endpoint]
if gemini_endpoint in API_URL_REDIRECT: gemini_endpoint = API_URL_REDIRECT[gemini_endpoint]
if claude_endpoint in API_URL_REDIRECT: claude_endpoint = API_URL_REDIRECT[claude_endpoint]
# 获取tokenizer # 获取tokenizer
tokenizer_gpt35 = LazyloadTiktoken("gpt-3.5-turbo") tokenizer_gpt35 = LazyloadTiktoken("gpt-3.5-turbo")
@ -291,7 +295,7 @@ model_info = {
"gemini-pro": { "gemini-pro": {
"fn_with_ui": genai_ui, "fn_with_ui": genai_ui,
"fn_without_ui": genai_noui, "fn_without_ui": genai_noui,
"endpoint": None, "endpoint": gemini_endpoint,
"max_token": 1024 * 32, "max_token": 1024 * 32,
"tokenizer": tokenizer_gpt35, "tokenizer": tokenizer_gpt35,
"token_cnt": get_token_num_gpt35, "token_cnt": get_token_num_gpt35,
@ -299,7 +303,7 @@ model_info = {
"gemini-pro-vision": { "gemini-pro-vision": {
"fn_with_ui": genai_ui, "fn_with_ui": genai_ui,
"fn_without_ui": genai_noui, "fn_without_ui": genai_noui,
"endpoint": None, "endpoint": gemini_endpoint,
"max_token": 1024 * 32, "max_token": 1024 * 32,
"tokenizer": tokenizer_gpt35, "tokenizer": tokenizer_gpt35,
"token_cnt": get_token_num_gpt35, "token_cnt": get_token_num_gpt35,
@ -349,25 +353,57 @@ for model in AVAIL_LLM_MODELS:
model_info.update({model: mi}) model_info.update({model: mi})
# -=-=-=-=-=-=- 以下部分是新加入的模型，可能附带额外依赖 -=-=-=-=-=-=- # -=-=-=-=-=-=- 以下部分是新加入的模型，可能附带额外依赖 -=-=-=-=-=-=-
if "claude-1-100k" in AVAIL_LLM_MODELS or "claude-2" in AVAIL_LLM_MODELS: # claude家族
claude_models = ["claude-instant-1.2","claude-2.0","claude-2.1","claude-3-sonnet-20240229","claude-3-opus-20240229"]
if any(item in claude_models for item in AVAIL_LLM_MODELS):
from .bridge_claude import predict_no_ui_long_connection as claude_noui from .bridge_claude import predict_no_ui_long_connection as claude_noui
from .bridge_claude import predict as claude_ui from .bridge_claude import predict as claude_ui
model_info.update({ model_info.update({
"claude-1-100k": { "claude-instant-1.2": {
"fn_with_ui": claude_ui, "fn_with_ui": claude_ui,
"fn_without_ui": claude_noui, "fn_without_ui": claude_noui,
"endpoint": None, "endpoint": claude_endpoint,
"max_token": 8196, "max_token": 100000,
"tokenizer": tokenizer_gpt35, "tokenizer": tokenizer_gpt35,
"token_cnt": get_token_num_gpt35, "token_cnt": get_token_num_gpt35,
}, },
}) })
model_info.update({ model_info.update({
"claude-2": { "claude-2.0": {
"fn_with_ui": claude_ui, "fn_with_ui": claude_ui,
"fn_without_ui": claude_noui, "fn_without_ui": claude_noui,
"endpoint": None, "endpoint": claude_endpoint,
"max_token": 8196, "max_token": 100000,
"tokenizer": tokenizer_gpt35,
"token_cnt": get_token_num_gpt35,
},
})
model_info.update({
"claude-2.1": {
"fn_with_ui": claude_ui,
"fn_without_ui": claude_noui,
"endpoint": claude_endpoint,
"max_token": 200000,
"tokenizer": tokenizer_gpt35,
"token_cnt": get_token_num_gpt35,
},
})
model_info.update({
"claude-3-sonnet-20240229": {
"fn_with_ui": claude_ui,
"fn_without_ui": claude_noui,
"endpoint": claude_endpoint,
"max_token": 200000,
"tokenizer": tokenizer_gpt35,
"token_cnt": get_token_num_gpt35,
},
})
model_info.update({
"claude-3-opus-20240229": {
"fn_with_ui": claude_ui,
"fn_without_ui": claude_noui,
"endpoint": claude_endpoint,
"max_token": 200000,
"tokenizer": tokenizer_gpt35, "tokenizer": tokenizer_gpt35,
"token_cnt": get_token_num_gpt35, "token_cnt": get_token_num_gpt35,
}, },
@ -675,22 +711,28 @@ if "deepseekcoder" in AVAIL_LLM_MODELS: # deepseekcoder
}) })
except: except:
print(trimmed_format_exc()) print(trimmed_format_exc())
# if "skylark" in AVAIL_LLM_MODELS: # -=-=-=-=-=-=- one-api 对齐支持 -=-=-=-=-=-=-
# try: for model in [m for m in AVAIL_LLM_MODELS if m.startswith("one-api-")]:
# from .bridge_skylark2 import predict_no_ui_long_connection as skylark_noui # 为了更灵活地接入one-api多模型管理界面，设计了此接口，例子：AVAIL_LLM_MODELS = ["one-api-mixtral-8x7b(max_token=6666)"]
# from .bridge_skylark2 import predict as skylark_ui # 其中
# model_info.update({ # "one-api-" 是前缀（必要）
# "skylark": { # "mixtral-8x7b" 是模型名（必要）
# "fn_with_ui": skylark_ui, # "(max_token=6666)" 是配置（非必要）
# "fn_without_ui": skylark_noui, try:
# "endpoint": None, _, max_token_tmp = read_one_api_model_name(model)
# "max_token": 4096, except:
# "tokenizer": tokenizer_gpt35, print(f"one-api模型 {model} 的 max_token 配置不是整数，请检查配置文件。")
# "token_cnt": get_token_num_gpt35, continue
# } model_info.update({
# }) model: {
# except: "fn_with_ui": chatgpt_ui,
# print(trimmed_format_exc()) "fn_without_ui": chatgpt_noui,
"endpoint": openai_endpoint,
"max_token": max_token_tmp,
"tokenizer": tokenizer_gpt35,
"token_cnt": get_token_num_gpt35,
},
})
# <-- 用于定义和切换多个azure模型 --> # <-- 用于定义和切换多个azure模型 -->

5
request_llms/bridge_chatgpt.py
View File

@ -21,7 +21,7 @@ import random
# config_private.py放自己的秘密如API和代理网址 # config_private.py放自己的秘密如API和代理网址
# 读取时首先看是否存在私密的config_private配置文件（不受git管控），如果有，则覆盖原config文件 # 读取时首先看是否存在私密的config_private配置文件（不受git管控），如果有，则覆盖原config文件
from toolbox import get_conf, update_ui, is_any_api_key, select_api_key, what_keys, clip_history, trimmed_format_exc, is_the_upload_folder from toolbox import get_conf, update_ui, is_any_api_key, select_api_key, what_keys, clip_history, trimmed_format_exc, is_the_upload_folder, read_one_api_model_name
proxies, TIMEOUT_SECONDS, MAX_RETRY, API_ORG, AZURE_CFG_ARRAY = \ proxies, TIMEOUT_SECONDS, MAX_RETRY, API_ORG, AZURE_CFG_ARRAY = \
get_conf('proxies', 'TIMEOUT_SECONDS', 'MAX_RETRY', 'API_ORG', 'AZURE_CFG_ARRAY') get_conf('proxies', 'TIMEOUT_SECONDS', 'MAX_RETRY', 'API_ORG', 'AZURE_CFG_ARRAY')
@ -358,6 +358,9 @@ def generate_payload(inputs, llm_kwargs, history, system_prompt, stream):
model = llm_kwargs['llm_model'] model = llm_kwargs['llm_model']
if llm_kwargs['llm_model'].startswith('api2d-'): if llm_kwargs['llm_model'].startswith('api2d-'):
model = llm_kwargs['llm_model'][len('api2d-'):] model = llm_kwargs['llm_model'][len('api2d-'):]
if llm_kwargs['llm_model'].startswith('one-api-'):
model = llm_kwargs['llm_model'][len('one-api-'):]
model, _ = read_one_api_model_name(model)
if model == "gpt-3.5-random": # 随机选择, 绕过openai访问频率限制 if model == "gpt-3.5-random": # 随机选择, 绕过openai访问频率限制
model = random.choice([ model = random.choice([

143
request_llms/bridge_claude.py
View File

@ -11,13 +11,12 @@
""" """
import os import os
import json
import time import time
import gradio as gr
import logging
import traceback import traceback
import requests from toolbox import get_conf, update_ui, trimmed_format_exc, encode_image, every_image_file_in_path
import importlib
picture_system_prompt = "\n当回复图像时,必须说明正在回复哪张图像。所有图像仅在最后一个问题中提供,即使它们在历史记录中被提及。请使用'这是第X张图像:'的格式来指明您正在描述的是哪张图像。"
Claude_3_Models = ["claude-3-sonnet-20240229", "claude-3-opus-20240229"]
# config_private.py放自己的秘密如API和代理网址 # config_private.py放自己的秘密如API和代理网址
# 读取时首先看是否存在私密的config_private配置文件（不受git管控），如果有，则覆盖原config文件 # 读取时首先看是否存在私密的config_private配置文件（不受git管控），如果有，则覆盖原config文件
@ -56,7 +55,8 @@ def predict_no_ui_long_connection(inputs, llm_kwargs, history=[], sys_prompt="",
""" """
from anthropic import Anthropic from anthropic import Anthropic
watch_dog_patience = 5 # 看门狗的耐心, 设置5秒即可 watch_dog_patience = 5 # 看门狗的耐心, 设置5秒即可
prompt = generate_payload(inputs, llm_kwargs, history, system_prompt=sys_prompt, stream=True) if inputs == "": inputs = "空空如也的输入栏"
message = generate_payload(inputs, llm_kwargs, history, stream=True, image_paths=None)
retry = 0 retry = 0
if len(ANTHROPIC_API_KEY) == 0: if len(ANTHROPIC_API_KEY) == 0:
raise RuntimeError("没有设置ANTHROPIC_API_KEY选项") raise RuntimeError("没有设置ANTHROPIC_API_KEY选项")
@ -65,15 +65,16 @@ def predict_no_ui_long_connection(inputs, llm_kwargs, history=[], sys_prompt="",
try: try:
# make a POST request to the API endpoint, stream=False # make a POST request to the API endpoint, stream=False
from .bridge_all import model_info from .bridge_all import model_info
anthropic = Anthropic(api_key=ANTHROPIC_API_KEY) anthropic = Anthropic(api_key=ANTHROPIC_API_KEY, base_url=model_info[llm_kwargs['llm_model']]['endpoint'])
# endpoint = model_info[llm_kwargs['llm_model']]['endpoint'] # endpoint = model_info[llm_kwargs['llm_model']]['endpoint']
# with ProxyNetworkActivate() # with ProxyNetworkActivate()
stream = anthropic.completions.create( stream = anthropic.messages.create(
prompt=prompt, messages=message,
max_tokens_to_sample=4096, # The maximum number of tokens to generate before stopping. max_tokens=4096, # The maximum number of tokens to generate before stopping.
model=llm_kwargs['llm_model'], model=llm_kwargs['llm_model'],
stream=True, stream=True,
temperature = llm_kwargs['temperature'] temperature = llm_kwargs['temperature'],
system=sys_prompt
) )
break break
except Exception as e: except Exception as e:
@ -84,11 +85,15 @@ def predict_no_ui_long_connection(inputs, llm_kwargs, history=[], sys_prompt="",
result = '' result = ''
try: try:
for completion in stream: for completion in stream:
result += completion.completion if completion.type == "message_start" or completion.type == "content_block_start":
if not console_slience: print(completion.completion, end='') continue
elif completion.type == "message_stop" or completion.type == "content_block_stop" or completion.type == "message_delta":
break
result += completion.delta.text
if not console_slience: print(completion.delta.text, end='')
if observe_window is not None: if observe_window is not None:
# 观测窗，把已经获取的数据显示出去 # 观测窗，把已经获取的数据显示出去
if len(observe_window) >= 1: observe_window[0] += completion.completion if len(observe_window) >= 1: observe_window[0] += completion.delta.text
# 看门狗，如果超过期限没有喂狗，则终止 # 看门狗，如果超过期限没有喂狗，则终止
if len(observe_window) >= 2: if len(observe_window) >= 2:
if (time.time()-observe_window[1]) > watch_dog_patience: if (time.time()-observe_window[1]) > watch_dog_patience:
@ -98,6 +103,10 @@ def predict_no_ui_long_connection(inputs, llm_kwargs, history=[], sys_prompt="",
return result return result
def make_media_input(history,inputs,image_paths):
for image_path in image_paths:
inputs = inputs + f'<br/><br/><div align="center"><img src="file={os.path.abspath(image_path)}"></div>'
return inputs
def predict(inputs, llm_kwargs, plugin_kwargs, chatbot, history=[], system_prompt='', stream = True, additional_fn=None): def predict(inputs, llm_kwargs, plugin_kwargs, chatbot, history=[], system_prompt='', stream = True, additional_fn=None):
""" """
@ -109,6 +118,7 @@ def predict(inputs, llm_kwargs, plugin_kwargs, chatbot, history=[], system_promp
chatbot 为WebUI中显示的对话列表，修改它，然后yeild出去，可以直接修改对话界面内容 chatbot 为WebUI中显示的对话列表，修改它，然后yeild出去，可以直接修改对话界面内容
additional_fn代表点击的哪个按钮，按钮见functional.py additional_fn代表点击的哪个按钮，按钮见functional.py
""" """
if inputs == "": inputs = "空空如也的输入栏"
from anthropic import Anthropic from anthropic import Anthropic
if len(ANTHROPIC_API_KEY) == 0: if len(ANTHROPIC_API_KEY) == 0:
chatbot.append((inputs, "没有设置ANTHROPIC_API_KEY")) chatbot.append((inputs, "没有设置ANTHROPIC_API_KEY"))
@ -119,13 +129,23 @@ def predict(inputs, llm_kwargs, plugin_kwargs, chatbot, history=[], system_promp
from core_functional import handle_core_functionality from core_functional import handle_core_functionality
inputs, history = handle_core_functionality(additional_fn, inputs, history, chatbot) inputs, history = handle_core_functionality(additional_fn, inputs, history, chatbot)
raw_input = inputs have_recent_file, image_paths = every_image_file_in_path(chatbot)
logging.info(f'[raw_input] {raw_input}') if len(image_paths) > 20:
chatbot.append((inputs, "")) chatbot.append((inputs, "图片数量超过api上限(20张)"))
yield from update_ui(chatbot=chatbot, history=history, msg="等待响应") # 刷新界面 yield from update_ui(chatbot=chatbot, history=history, msg="等待响应")
return
if any([llm_kwargs['llm_model'] == model for model in Claude_3_Models]) and have_recent_file:
if inputs == "" or inputs == "空空如也的输入栏": inputs = "请描述给出的图片"
system_prompt += picture_system_prompt # 由于没有单独的参数保存包含图片的历史，所以只能通过提示词对第几张图片进行定位
chatbot.append((make_media_input(history,inputs, image_paths), ""))
yield from update_ui(chatbot=chatbot, history=history, msg="等待响应") # 刷新界面
else:
chatbot.append((inputs, ""))
yield from update_ui(chatbot=chatbot, history=history, msg="等待响应") # 刷新界面
try: try:
prompt = generate_payload(inputs, llm_kwargs, history, system_prompt, stream) message = generate_payload(inputs, llm_kwargs, history, stream, image_paths)
except RuntimeError as e: except RuntimeError as e:
chatbot[-1] = (inputs, f"您提供的api-key不满足要求，不包含任何可用于{llm_kwargs['llm_model']}的api-key。您可能选择了错误的模型或请求源。") chatbot[-1] = (inputs, f"您提供的api-key不满足要求，不包含任何可用于{llm_kwargs['llm_model']}的api-key。您可能选择了错误的模型或请求源。")
yield from update_ui(chatbot=chatbot, history=history, msg="api-key不满足要求") # 刷新界面 yield from update_ui(chatbot=chatbot, history=history, msg="api-key不满足要求") # 刷新界面
@ -138,17 +158,17 @@ def predict(inputs, llm_kwargs, plugin_kwargs, chatbot, history=[], system_promp
try: try:
# make a POST request to the API endpoint, stream=True # make a POST request to the API endpoint, stream=True
from .bridge_all import model_info from .bridge_all import model_info
anthropic = Anthropic(api_key=ANTHROPIC_API_KEY) anthropic = Anthropic(api_key=ANTHROPIC_API_KEY, base_url=model_info[llm_kwargs['llm_model']]['endpoint'])
# endpoint = model_info[llm_kwargs['llm_model']]['endpoint'] # endpoint = model_info[llm_kwargs['llm_model']]['endpoint']
# with ProxyNetworkActivate() # with ProxyNetworkActivate()
stream = anthropic.completions.create( stream = anthropic.messages.create(
prompt=prompt, messages=message,
max_tokens_to_sample=4096, # The maximum number of tokens to generate before stopping. max_tokens=4096, # The maximum number of tokens to generate before stopping.
model=llm_kwargs['llm_model'], model=llm_kwargs['llm_model'],
stream=True, stream=True,
temperature = llm_kwargs['temperature'] temperature = llm_kwargs['temperature'],
system=system_prompt
) )
break break
except: except:
retry += 1 retry += 1
@ -160,8 +180,12 @@ def predict(inputs, llm_kwargs, plugin_kwargs, chatbot, history=[], system_promp
gpt_replying_buffer = "" gpt_replying_buffer = ""
for completion in stream: for completion in stream:
if completion.type == "message_start" or completion.type == "content_block_start":
continue
elif completion.type == "message_stop" or completion.type == "content_block_stop" or completion.type == "message_delta":
break
try: try:
gpt_replying_buffer = gpt_replying_buffer + completion.completion gpt_replying_buffer = gpt_replying_buffer + completion.delta.text
history[-1] = gpt_replying_buffer history[-1] = gpt_replying_buffer
chatbot[-1] = (history[-2], history[-1]) chatbot[-1] = (history[-2], history[-1])
yield from update_ui(chatbot=chatbot, history=history, msg='正常') # 刷新界面 yield from update_ui(chatbot=chatbot, history=history, msg='正常') # 刷新界面
@ -173,56 +197,51 @@ def predict(inputs, llm_kwargs, plugin_kwargs, chatbot, history=[], system_promp
yield from update_ui(chatbot=chatbot, history=history, msg="Json异常" + tb_str) # 刷新界面 yield from update_ui(chatbot=chatbot, history=history, msg="Json异常" + tb_str) # 刷新界面
return return
def generate_payload(inputs, llm_kwargs, history, stream, image_paths):
# https://github.com/jtsang4/claude-to-chatgpt/blob/main/claude_to_chatgpt/adapter.py
def convert_messages_to_prompt(messages):
prompt = ""
role_map = {
"system": "Human",
"user": "Human",
"assistant": "Assistant",
}
for message in messages:
role = message["role"]
content = message["content"]
transformed_role = role_map[role]
prompt += f"\n\n{transformed_role.capitalize()}: {content}"
prompt += "\n\nAssistant: "
return prompt
def generate_payload(inputs, llm_kwargs, history, system_prompt, stream):
""" """
整合所有信息，选择LLM模型，生成http请求，为发送请求做准备 整合所有信息，选择LLM模型，生成http请求，为发送请求做准备
""" """
from anthropic import Anthropic, HUMAN_PROMPT, AI_PROMPT
conversation_cnt = len(history) // 2 conversation_cnt = len(history) // 2
messages = [{"role": "system", "content": system_prompt}] messages = []
if conversation_cnt: if conversation_cnt:
for index in range(0, 2*conversation_cnt, 2): for index in range(0, 2*conversation_cnt, 2):
what_i_have_asked = {} what_i_have_asked = {}
what_i_have_asked["role"] = "user" what_i_have_asked["role"] = "user"
what_i_have_asked["content"] = history[index] what_i_have_asked["content"] = [{"type": "text", "text": history[index]}]
what_gpt_answer = {} what_gpt_answer = {}
what_gpt_answer["role"] = "assistant" what_gpt_answer["role"] = "assistant"
what_gpt_answer["content"] = history[index+1] what_gpt_answer["content"] = [{"type": "text", "text": history[index+1]}]
if what_i_have_asked["content"] != "": if what_i_have_asked["content"][0]["text"] != "":
if what_gpt_answer["content"] == "": continue if what_i_have_asked["content"][0]["text"] == "": continue
if what_gpt_answer["content"] == timeout_bot_msg: continue if what_i_have_asked["content"][0]["text"] == timeout_bot_msg: continue
messages.append(what_i_have_asked) messages.append(what_i_have_asked)
messages.append(what_gpt_answer) messages.append(what_gpt_answer)
else: else:
messages[-1]['content'] = what_gpt_answer['content'] messages[-1]['content'][0]['text'] = what_gpt_answer['content'][0]['text']
what_i_ask_now = {} if any([llm_kwargs['llm_model'] == model for model in Claude_3_Models]) and image_paths:
what_i_ask_now["role"] = "user" base64_images = []
what_i_ask_now["content"] = inputs for image_path in image_paths:
base64_images.append(encode_image(image_path))
what_i_ask_now = {}
what_i_ask_now["role"] = "user"
what_i_ask_now["content"] = []
for base64_image in base64_images:
what_i_ask_now["content"].append({
"type": "image",
"source": {
"type": "base64",
"media_type": "image/jpeg",
"data": base64_image,
}
})
what_i_ask_now["content"].append({"type": "text", "text": inputs})
else:
what_i_ask_now = {}
what_i_ask_now["role"] = "user"
what_i_ask_now["content"] = [{"type": "text", "text": inputs}]
messages.append(what_i_ask_now) messages.append(what_i_ask_now)
prompt = convert_messages_to_prompt(messages) return messages
return prompt

4
request_llms/bridge_google_gemini.py
View File

@ -20,7 +20,7 @@ def predict_no_ui_long_connection(inputs, llm_kwargs, history=[], sys_prompt="",
if get_conf("GEMINI_API_KEY") == "": if get_conf("GEMINI_API_KEY") == "":
raise ValueError(f"请配置 GEMINI_API_KEY。") raise ValueError(f"请配置 GEMINI_API_KEY。")
genai = GoogleChatInit() genai = GoogleChatInit(llm_kwargs)
watch_dog_patience = 5 # 看门狗的耐心, 设置5秒即可 watch_dog_patience = 5 # 看门狗的耐心, 设置5秒即可
gpt_replying_buffer = '' gpt_replying_buffer = ''
stream_response = genai.generate_chat(inputs, llm_kwargs, history, sys_prompt) stream_response = genai.generate_chat(inputs, llm_kwargs, history, sys_prompt)
@ -70,7 +70,7 @@ def predict(inputs, llm_kwargs, plugin_kwargs, chatbot, history=[], system_promp
chatbot.append((inputs, "")) chatbot.append((inputs, ""))
yield from update_ui(chatbot=chatbot, history=history) yield from update_ui(chatbot=chatbot, history=history)
genai = GoogleChatInit() genai = GoogleChatInit(llm_kwargs)
retry = 0 retry = 0
while True: while True:
try: try:

6
request_llms/com_google.py
View File

@ -114,8 +114,10 @@ def html_local_img(__file, layout="left", max_width=None, max_height=None, md=Tr
class GoogleChatInit: class GoogleChatInit:
def __init__(self): def __init__(self, llm_kwargs):
self.url_gemini = "https://generativelanguage.googleapis.com/v1beta/models/%m:streamGenerateContent?key=%k" from .bridge_all import model_info
endpoint = model_info[llm_kwargs['llm_model']]['endpoint']
self.url_gemini = endpoint + "/%m:streamGenerateContent?key=%k"
def generate_chat(self, inputs, llm_kwargs, history, system_prompt): def generate_chat(self, inputs, llm_kwargs, history, system_prompt):
headers, payload = self.generate_message_payload( headers, payload = self.generate_message_payload(

2
requirements.txt
View File

@ -8,6 +8,7 @@ pydantic==2.5.2
protobuf==3.18 protobuf==3.18
transformers>=4.27.1 transformers>=4.27.1
scipdf_parser>=0.52 scipdf_parser>=0.52
anthropic>=0.18.1
python-markdown-math python-markdown-math
pymdown-extensions pymdown-extensions
websocket-client websocket-client
@ -16,7 +17,6 @@ prompt_toolkit
latex2mathml latex2mathml
python-docx python-docx
mdtex2html mdtex2html
anthropic
pyautogen pyautogen
colorama colorama
Markdown Markdown

2
shared_utils/key_pattern_manager.py
View File

@ -62,7 +62,7 @@ def select_api_key(keys, llm_model):
avail_key_list = [] avail_key_list = []
key_list = keys.split(',') key_list = keys.split(',')
if llm_model.startswith('gpt-'): if llm_model.startswith('gpt-') or llm_model.startswith('one-api-'):
for k in key_list: for k in key_list:
if is_openai_api_key(k): avail_key_list.append(k) if is_openai_api_key(k): avail_key_list.append(k)

34
shared_utils/map_names.py Normal file
View File

@ -0,0 +1,34 @@
import re
mapping_dic = {
# "qianfan": "qianfan（文心一言大模型）",
# "zhipuai": "zhipuai（智谱GLM4超级模型🔥）",
# "gpt-4-1106-preview": "gpt-4-1106-preview（新调优版本GPT-4🔥）",
# "gpt-4-vision-preview": "gpt-4-vision-preview（识图模型GPT-4V）",
}
rev_mapping_dic = {}
for k, v in mapping_dic.items():
rev_mapping_dic[v] = k
def map_model_to_friendly_names(m):
if m in mapping_dic:
return mapping_dic[m]
return m
def map_friendly_names_to_model(m):
if m in rev_mapping_dic:
return rev_mapping_dic[m]
return m
def read_one_api_model_name(model: str):
"""return real model name and max_token.
"""
max_token_pattern = r"\(max_token=(\d+)\)"
match = re.search(max_token_pattern, model)
if match:
max_token_tmp = match.group(1) # 获取 max_token 的值
max_token_tmp = int(max_token_tmp)
model = re.sub(max_token_pattern, "", model) # 从原字符串中删除 "(max_token=...)"
else:
max_token_tmp = 4096
return model, max_token_tmp

65
toolbox.py
View File

@ -25,6 +25,9 @@ from shared_utils.text_mask import apply_gpt_academic_string_mask
from shared_utils.text_mask import build_gpt_academic_masked_string from shared_utils.text_mask import build_gpt_academic_masked_string
from shared_utils.text_mask import apply_gpt_academic_string_mask_langbased from shared_utils.text_mask import apply_gpt_academic_string_mask_langbased
from shared_utils.text_mask import build_gpt_academic_masked_string_langbased from shared_utils.text_mask import build_gpt_academic_masked_string_langbased
from shared_utils.map_names import map_friendly_names_to_model
from shared_utils.map_names import map_model_to_friendly_names
from shared_utils.map_names import read_one_api_model_name
from shared_utils.handle_upload import html_local_file from shared_utils.handle_upload import html_local_file
from shared_utils.handle_upload import html_local_img from shared_utils.handle_upload import html_local_img
from shared_utils.handle_upload import file_manifest_filter_type from shared_utils.handle_upload import file_manifest_filter_type
@ -919,6 +922,18 @@ def have_any_recent_upload_image_files(chatbot):
else: else:
return False, None # most_recent_uploaded is too old return False, None # most_recent_uploaded is too old
# Claude3 model supports graphic context dialogue, reads all images
def every_image_file_in_path(chatbot):
if chatbot is None:
return False, [] # chatbot is None
most_recent_uploaded = chatbot._cookies.get("most_recent_uploaded", None)
if not most_recent_uploaded:
return False, [] # most_recent_uploaded is None
path = most_recent_uploaded["path"]
file_manifest = get_pictures_list(path)
if len(file_manifest) == 0:
return False, []
return True, file_manifest
# Function to encode the image # Function to encode the image
def encode_image(image_path): def encode_image(image_path):
@ -939,3 +954,53 @@ def check_packages(packages=[]):
spam_spec = importlib.util.find_spec(p) spam_spec = importlib.util.find_spec(p)
if spam_spec is None: if spam_spec is None:
raise ModuleNotFoundError raise ModuleNotFoundError
def map_file_to_sha256(file_path):
import hashlib
with open(file_path, 'rb') as file:
content = file.read()
# Calculate the SHA-256 hash of the file contents
sha_hash = hashlib.sha256(content).hexdigest()
return sha_hash
def check_repeat_upload(new_pdf_path, pdf_hash):
'''
检查历史上传的文件是否与新上传的文件相同，如果相同则返回(True, 重复文件路径)，否则返回(False，None)
'''
from toolbox import get_conf
import PyPDF2
user_upload_dir = os.path.dirname(os.path.dirname(new_pdf_path))
file_name = os.path.basename(new_pdf_path)
file_manifest = [f for f in glob.glob(f'{user_upload_dir}/**/{file_name}', recursive=True)]
for saved_file in file_manifest:
with open(new_pdf_path, 'rb') as file1, open(saved_file, 'rb') as file2:
reader1 = PyPDF2.PdfFileReader(file1)
reader2 = PyPDF2.PdfFileReader(file2)
# 比较页数是否相同
if reader1.getNumPages() != reader2.getNumPages():
continue
# 比较每一页的内容是否相同
for page_num in range(reader1.getNumPages()):
page1 = reader1.getPage(page_num).extractText()
page2 = reader2.getPage(page_num).extractText()
if page1 != page2:
continue
maybe_project_dir = glob.glob('{}/**/{}'.format(get_log_folder(), pdf_hash + ".tag"), recursive=True)
if len(maybe_project_dir) > 0:
return True, os.path.dirname(maybe_project_dir[0])
# 如果所有页的内容都相同，返回 True
return False, None

4
version
View File

@ -1,5 +1,5 @@
{ {
"version": 3.72, "version": 3.73,
"show_feature": true, "show_feature": true,
"new_feature": "支持切换多个智谱ai模型 <-> 用绘图功能增强部分插件 <-> 基础功能区支持自动切换中英提示词 <-> 支持Mermaid绘图库（让大模型绘制脑图） <-> 支持Gemini-pro <-> 支持直接拖拽文件到上传区 <-> 支持将图片粘贴到输入区" "new_feature": "优化oneapi接入方法 <-> 接入月之暗面模型 <-> 支持切换多个智谱ai模型 <-> 用绘图功能增强部分插件 <-> 基础功能区支持自动切换中英提示词 <-> 支持Mermaid绘图库（让大模型绘制脑图）"
} }