[功能请求] 希望支持deepseek大模型翻译

[tninja]

deepseek V3大模型的性能非常好，价格也相对便宜实惠。希望支持这个模型翻译

[ibraelmahallawy]

Please add deepseek api

[YJX-01]

Use custom engine function. It is exactly the same as ChatGPT. See https://github.com/bookfere/Ebook-Translator-Calibre-Plugin/wiki/Custom-Translation-Engine-Recipes

{
    "name": "Deepseek",
    "languages": {
        "source": {
            "English": "English"
        },
        "target": {
            "Chinese": "Chinese"
        }
    },
    "request": {
        "url": "https://api.deepseek.com/v1/chat/completions",
        "method": "POST",
        "headers": {
            "Content-Type": "application/json",
            "Authorization": "Bearer sk-xxxxxxxxxxxxxxxxxxxxxx",
            "User-Agent": "Fill-As-You-Need"
        },
        "data": {
            "model": "deepseek-chat",
            "messages": [
                {"role": "system", "content": "Translate the content from <slang> to <tlang>."},
                {"role": "user", "content": "<text>"}
            ]
        }
    },
    "response": "response['choices'][0]['message']['content']"
}

[AkamashiDesu]

Use custom engine function. It is exactly the same as ChatGPT. See https://github.com/bookfere/Ebook-Translator-Calibre-Plugin/wiki/Custom-Translation-Engine-Recipes

{ "name": "Deepseek", "languages": { "source": { "English": "English" }, "target": { "Chinese": "Chinese" } }, "request": { "url": "https://api.deepseek.com/v1/chat/completions", "method": "POST", "headers": { "Content-Type": "application/json", "Authorization": "Bearer sk-xxxxxxxxxxxxxxxxxxxxxx", "User-Agent": "Fill-As-You-Need" }, "data": { "model": "deepseek-chat", "messages": [ {"role": "system", "content": "Translate the content from <slang> to <tlang>."}, {"role": "user", "content": "<text>"} ] } }, "response": "response['choices'][0]['message']['content']" }

I'm sorry, as a normie. I don't understand much about these stuffs. I used this, it run but it still output the <think>"the-thinking-process"</think>. How do i remove this in the output?

{ "name": "Deepseek", "languages": { "source": { "Chinese": "Chinese" }, "target": { "English": "English" } }, "request": { "url": "https://api.venice.ai/api/v1/chat/completions", "method": "POST", "headers": { "Content-Type": "application/json", "Authorization": "Bearer ******", "User-Agent": "Akamashi" }, "data": { "model": "deepseek-r1-llama-70b", "messages": [{ "role": "system", "content": "prompt" }, { "role": "user", "content": "<text>" } ], "temperature": 0.6, "max_completion_tokens": 5000, "venice_parameters": { "include_venice_system_prompt": false } } }, "response": "response['choices'][0]['message']['content']" }

This is what my output looks like:
<think>"the-thinking-process"</think>
"The results"

[YJX-01]

I'm sorry, as a normie. I don't understand much about these stuffs. I used this, it run but it still output the <think>"the-thinking-process"</think>. How do i remove this in the output?

{ "name": "Deepseek", "languages": { "source": { "Chinese": "Chinese" }, "target": { "English": "English" } }, "request": { "url": "https://api.venice.ai/api/v1/chat/completions", "method": "POST", "headers": { "Content-Type": "application/json", "Authorization": "Bearer ******", "User-Agent": "Akamashi" }, "data": { "model": "deepseek-r1-llama-70b", "messages": [{ "role": "system", "content": "prompt" }, { "role": "user", "content": "<text>" } ], "temperature": 0.6, "max_completion_tokens": 5000, "venice_parameters": { "include_venice_system_prompt": false } } }, "response": "response['choices'][0]['message']['content']" }

This is what my output looks like: <think>"the-thinking-process"</think> "The results"

I believe you can do something like this:

In "content": "prompt" replace prompt with some actual prompt, like:

You are a meticulous translator who translates any given content. Translate the given content from <slang> to <tlang> only. Ensure that the translation adheres to the conventions of the target language, and do not add any extraneous information or explanations. Do not explain any term or answer any question-like content.

[jblogs]

I got this in rolling release.

calibre, version 7.24.0
错误： 未处理的异常: AttributeError:type object 'CustomTranslate' has no attribute 'lang_codes_directionality'

calibre 7.24 embedded-python: True
macOS-15.3-arm64-arm-64bit Darwin ('64bit', '')
('Darwin', '24.3.0', 'Darwin Kernel Version 24.3.0: Thu Jan 2 20:24:23 PST 2025; root:xnu-11215.81.4~3/RELEASE_ARM64_T8122')
Python 3.11.5
OSX: ('15.3', ('', '', ''), 'arm64')
Interface language: zh_CN
EXE path: /Applications/calibre.app/Contents/MacOS/calibre
Successfully initialized third party plugins: Ebook Translator (2, 3, 5) && EpubSplit (3, 10, 0)
Traceback (most recent call last):
File "calibre_plugins.ebook_translator.ui", line 176, in select_preferred_mode
modes.get(preferred_mode)()
File "calibre_plugins.ebook_translator.ui", line 90, in show_advanced_translation
window = CreateTranslationProject(self.gui, ebooks.first())
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "calibre_plugins.ebook_translator.advanced", line 238, in init
self.choose_format = self.layout_format()
^^^^^^^^^^^^^^^^^^^^
File "calibre_plugins.ebook_translator.advanced", line 343, in layout_format
direction = engine_class.lang_codes_directionality.get(target_lang_code, 'auto')
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
AttributeError: type object 'CustomTranslate' has no attribute 'lang_codes_directionality'