详细内容
AI-Powered Subtitle Translation
GPT Subtitler
AI-Powered Subtitling: Fast, Accurate, Effortless.
Turn your videos into global content with precise,
context-aware translations
Try NowLearn More
Context-Aware Translation Excellence
Leveraging cutting-edge large language models to understand the meaning behind words, delivering translations that capture nuance, idioms, and cultural references.
Translation Comparison
Example 1 - Spanish to English
Context: Spanish idiom for being distracted or daydreaming
Original Text
(Spanish)
Estar en las nubes no me ayuda a pasar el examen.
Being distracted or daydreaming won't help me pass the exam
AI Translation
(Spanish to English)
Being daydreaming won't help me pass the exam.
Machine Translation
(Spanish to English)
Being in the clouds doesn't help me pass the exam.
GPT Subtitler understands context, cultural nuances, and idiomatic expressions for more natural translations.
Powerful Translation Features
Experience state-of-the-art subtitle translation with our comprehensive feature set.
Audio Transcription
Convert audio and video files directly to accurate subtitles using Whisper AI, with support for multiple languages.
100+ Languages
Translate your subtitles to and from over 100 languages with high accuracy.
Advanced Settings
Fine-tune your translations with customizable settings and preferences for different content types and audiences.
Batch Translation
Process multiple subtitle files simultaneously with consistent terminology and style across your entire content library.
Supported LLM Providers
Choose your preferred LLM provider to translate your subtitles.
OpenAI
Claude
Gemini
DeepSeek
Grok
OpenAI Compatible
How to Use GPT Subtitler
Complete your subtitle translations in just four simple steps.
1
Upload Subtitle File
Upload your subtitle file in any supported format to begin the translation process.
2
Select Translation Options
Choose your target language and adjust translation settings as needed.
3
Process Translation
Our AI models will translate your subtitles while preserving timing and formatting.
4
Download Translated Subtitles
Download your translated subtitles in your preferred format.
Try It Now
Frequently Asked Questions
Find answers to common questions about GPT Subtitler.
How do I translate subtitles?
- Sign in to your account or create a new account if you don't have one.
- Click on "Translate a new subtitle" to create a new translation project.
- Upload your subtitle file in the SRT format or paste the subtitle content directly into the editor.
- Click on "Create" to start the translation process.
- If you have an OpenAI/Claude/Gemini API key, you can enter it in the settings. Otherwise, you need to use your account token for translations.
- Configure the translation settings, such as source and target languages, start index, prompt, batch size and model.
- Click on "Translate" to begin the translation.
- The translation process will start, and you can monitor the progress in real-time. The translated subtitles will be displayed in the "Translation Messages" section.
- You can stop the translation process at any time and resume it later using the "start index" option.
- Once the translation is complete, you can download the translated subtitles in your desired format.
What file formats are supported for subtitle uploads?
GPT Subtitler supports the SRT (SubRip Text) file format for subtitle uploads. If your subtitles are in a different format, you can convert your subtitles file at subtitletools .
How much does it cost to use GPT Subtitler?
GPT Subtitler offers different pricing plans to suit your needs. We have a free subscription option, where you can receive a certain amount of Tokens daily, which can be used to translate a ~30 minutes video using the GPT-4o-mini model. For higher usage and advanced features, we offer Basic and Premium subscription plans. Please visit our pricing page for more details on the available plans and their pricing.
What do the different error codes mean and how do I fix them?
Token and Usage Related
Insufficient Tokens
Error: Insufficient tokens.
Please purchase more tokens or upgrade your subscription.
- Solution: Purchase more tokens, reduce batch size or use a cheaper model like gpt-4o-mini
- Prevention: Enable token estimation before translation to check costs
Usage Limit Exceeded
Error: Daily subscription usage limit exceede.
Please try again later or use your account Token to translate.
- Solution: Wait for daily subscription usage to reset, or use account tokens instead
- Prevention: Monitor your usage in settings page
No API Key or Subscription
Error: Subscription tier is free,
user has chosen not to use tokens, and no API key provided.
Please upgrade your subscription or choose to use tokens.
- Solution: Provide API key, use tokens, or upgrade subscription
- Prevention: Check translation settings before starting
Model and API Related
Maximum Tokens Reached
Error: Maximum tokens reached. Unable to parse the response
Please reduce the batch size in the advanced settings and try again.
- Solution: Reduce batch size in translation settings or disable Reflection or reasoning Mode
- Prevention: Use models with larger Max Generation Length for long batches
Parse Error
SyntaxError: Unable to parse JSON response.
Please retry using the "Fill Missing" feature. This may be due to the model outputting an incorrect format, which is an occasional error.
If the problem persists, try reducing the batch size or using a different model
- Solution: The model may have hallucinated, use "Fill Missing" feature to retry failed sections, or use a different model
- Prevention: Try reducing batch size or switching to a different model, or enable and generate a few-shot example using the few-shot example generator
Invalid API Key
Please provide a valid API key based on the selected model.
- Solution: Check your API key and ensure it matches the selected model
- Prevention: Verify your API key before starting translation
Invalid Model Selected
Invalid model selected, please choose a valid model
- Solution: Choose another available model from the dropdown list
- Prevention: If you need to use this specific model, please contact support
API Key Daily Limit Reached
Error: API key daily usage limit reached.
Please upgrade your subscription or try again tomorrow.
- Solution: Upgrade your subscription to avoid this limit, or wait until tomorrow
- Prevention: Consider upgrading your subscription to avoid daily limits
Unknown Error
An unknown error occurred while processing the current batch.
- Solution: Please try again later or switch to different model
- Prevention: N/A (server-side issue)
Concurrent Translation Related
Concurrency Limit Exceeded
The selected concurrency limit is too high for your subscription tier. Please reduce the concurrency limit.
- Solution: Reduce the concurrent batch limit in translation settings or upgrade your subscription tier for higher concurrency.
- Prevention: Check the maximum concurrency limit for your tier before setting a higher value.
Model Not Compatible With Concurrency
The selected model is not compatible with concurrent translation. Please select a different model or use your own API key to translate, or disable concurrent translation.
- Solution: Switch to a different model that supports concurrent processing or disable concurrent translation in the settings.
- Prevention: For models with rate limitations, use sequential translation mode instead of concurrent.
Request Data Not Found
Concurrent translation request data not found.
- Solution: Try refreshing the page and initiating the translation again.
- Prevention: Ensure you have a stable internet connection when starting large translation jobs.
Concurrent Translation Failed
Failed to process concurrent translation.
- Solution: Check for specific batch errors in the translation progress panel. You can try reducing the concurrency limit or switching to sequential mode.
- Prevention: For complex translations or when experiencing errors, try reducing the concurrency limit.
OpenAI Specific
Invalid OpenAI Request
Invalid OpenAI request. Please check the prompt and try again, it must contain the word 'json' or it might have exceed the maximum token limit of this model.
- Solution: Check if prompt contains 'json' keyword, or reduce batch size
- Prevention: Use default prompts or validated custom prompts
OpenAI Content Filtering
OpenAI content filtering detected potentially inappropriate content. Try switching to Gemini, Claude, or other providers, or modify your content to avoid policy violations.
- Solution: Switch to Gemini, Claude, or other providers that have different content policies, or modify your content to avoid policy violations
- Prevention: Review content for potentially sensitive material before translation
OpenAI Permission Denied
Your OpenAI API key does not have permission to use the specified resource.
- Solution: Check API key permissions in OpenAI dashboard
- Prevention: Ensure API key has correct access levels
Incorrect OpenAI API Key
Incorrect OpenAI API key provided, or you must be a member of an organization to use the API.
- Solution: Verify API key or ensure organization membership
- Prevention: Double-check API key format and validity
OpenAI Rate Limit Reached
OpenAI request rate limit reached or you have exceeded your current quota. Please try again later or check your plan and billing details.
- Solution: Wait a few minutes and try again, or switch to different model
- Prevention: Use paid API key or reduce concurrent translations
OpenAI Server Error
The OpenAI server encountered an error while processing your request. Please try again later.
- Solution: Wait and retry later or switch to different model
- Prevention: N/A (server-side issue)
OpenAI Services Overloaded
OpenAI's API is temporarily overloaded.
- Solution: Wait and retry later or switch to different model
- Prevention: N/A (server-side issue)
Claude Specific
Invalid Claude Request
Invalid Claude request. Please check the prompt and try again. Note that Claude only supports temperature between 0 and 1.0
- Solution: Check prompt format and translation settings, note that Claude only supports temperature between 0 and 1.0
- Prevention: Use default prompts or validated custom prompts
Claude Permission Denied
Your Claude API key does not have permission to use the specified resource.
- Solution: Verify API key permissions
- Prevention: Ensure correct API key access levels
Incorrect Claude API Key
Incorrect Claude API key provided, or you must be a member of an organization to use the API.
- Solution: Check and update API key
- Prevention: Verify API key before starting
Claude Rate Limit Reached
Claude request rate limit reached or you have exceeded your current quota. Please try again later or check your plan and billing details.
- Solution: Wait and retry later or switch to different model
- Prevention: Space out translation requests
Claude Server Error
Anthropic's server encountered an error while processing your request. Please try again later.
- Solution: Wait and retry later or switch to different model
- Prevention: N/A (server-side issue)
Claude Services Overloaded
Anthropic's API is temporarily overloaded.
- Solution: Wait and retry later or switch to different model
- Prevention: N/A (server-side issue)
Gemini Specific
Invalid Gemini Request
The request body to Gemini is malformed. Your API key may not be valid or the request is missing required parameters.
- Solution: Check API key and request parameters
- Prevention: Verify settings before starting
Gemini Permission Denied
Your API key doesn't have the required Gemini permissions. Check that your API key is set and has the right access.
- Solution: Check API key permissions
- Prevention: Ensure correct API key setup
Gemini Resource Not Found
The requested Gemini resource wasn't found.
- Solution: Verify model availability
- Prevention: Use supported models
Gemini Rate Limit Reached
Gemini request rate limit reached. Please try again later, use other models, or consider using your own Gemini API key. You can apply for a free key from Google at https://aistudio.google.com/app/apikey.
- Solution: Try again later or use different model
- Prevention: For free Gemini models, there is a daily request limit shared by all users, if too many people use it, you will be more likely to encounter this error. For paid Gemini models, if they are experimental models, they are more likely to encounter this error because they are still unstable versions. It is recommended to use stable versions (without -exp in the model name), and if needed, you can get a free Gemini API key at https://aistudio.google.com/app/apikey
Gemini Server Error
An unexpected error occurred on Google's side. Wait a bit and retry your request or use other models.
- Solution: Check your API key and wait and retry later, or switch to different model
- Prevention: N/A (server-side issue)
Gemini Service Unavailable
The Gemini service may be temporarily overloaded or down. Wait a bit and retry your request or use other models.
- Solution: Wait and retry later or switch to different model
- Prevention: N/A (server-side issue)
Gemini Server Error
An error occurred while fetching data from the Gemini API. Please try again later.
- Solution: Please try again later or switch to a different model
- Prevention: N/A (server-side issue)
Gemini Blocked by Safety
The content of the file is blocked by Gemini due to [SAFETY] reasons. Please try again with a different models or modify the content of the file.
- Solution: Please try again later or switch to a different model
- Prevention: N/A (server-side issue)
Gemini Blocked by Other Reasons
The content of the file is blocked by Gemini for [OTHER] reasons. Please try again with a different models or modify the content of the file.
- Solution: Please try again later or switch to a different model
- Prevention: N/A (server-side issue)
Gemini Blocked Prohibited Content
The content of the file is blocked by Gemini due to prohibited content detection. Please try again with a different model or modify the content of the file.
- Solution: Review and modify the content to comply with usage policies, or try a different model
- Prevention: Ensure content follows platform guidelines and usage policies
Gemini Fetch Error
An error occurred while fetching data from the Gemini API.
- Solution: Please try again later or switch to a different model
- Prevention: N/A (server-side issue)
DeepSeek Specific
Invalid DeepSeek Request
Invalid DeepSeek request format. Please modify your request body according to the error hints.
- Solution: Check request format
- Prevention: Use default settings
DeepSeek Authentication Failed
DeepSeek authentication failed. Please check your API key.
- Solution: Verify API key
- Prevention: Check API key validity
Insufficient DeepSeek Balance
Insufficient DeepSeek account balance. Please check your account balance and top up.
- Solution: Top up account
- Prevention: Monitor account balance
DeepSeek Rate Limit Reached
DeepSeek rate limit reached. Please pace your requests reasonably or temporarily switch to other LLM service providers.
- Solution: Wait and retry or switch to different model
- Prevention: Try use a larger batch size
Mistral Specific
Invalid Mistral Request
Invalid request to Mistral API
- Solution: Check request parameters
- Prevention: Use default settings
Mistral Permission Denied
Permission denied for Mistral API
- Solution: Check API key permissions
- Prevention: Verify API key setup
Mistral Rate Limit Reached
Mistral API rate limit reached
- Solution: Wait and retry or switch to different model
- Prevention: Try use a larger batch size
XAI (Grok) Specific
Invalid XAI Request
Invalid xAI request. Please check the prompt and try again.
- Solution: Check your API key format and validity
- Prevention: Verify your API key before starting translation
XAI Permission Denied
Your xAI API key does not have permission to use the specified resource.
- Solution: Verify API key permissions
- Prevention: Check API key access
XAI Rate Limit Reached
xAI request rate limit reached. Please try again later.
- Solution: Wait and retry or switch to different model
- Prevention: Try use a larger batch size
Qwen Specific
Invalid Qwen Request
Invalid Qwen request. Please check the prompt and try again.
- Solution: Check prompt format and translation settings
- Prevention: Use default prompts or validated custom prompts
Qwen Permission Denied
Your Qwen API key does not have permission to use the specified resource.
- Solution: Verify API key permissions
- Prevention: Ensure correct API key access levels
Qwen Rate Limit Reached
Qwen request rate limit reached. Please try again later.
- Solution: Wait and retry later or switch to different model
- Prevention: Space out translation requests
Qwen Server Error
The Qwen server encountered an error while processing your request.
- Solution: Wait and retry later or switch to different model
- Prevention: N/A (server-side issue)
Qwen Services Overloaded
Qwen's API is temporarily overloaded.
- Solution: Wait and retry later or switch to different model
- Prevention: N/A (server-side issue)
Third-Party Model Specific
Missing Third-Party Parameters
Error: Missing parameters for third-party model, please provide the API key, base url and the model name.
- Solution: Provide API key, base URL, and model name
- Prevention: Check all required fields are filled
Third-Party Authentication Failed
Third-party API authentication failed. Please check your API key.
- Solution: Check API key
- Prevention: Verify API key validity
Connection Related
Timeout Warning
Warning: Translation request has exceeded the maximum time limit.
Please try again later or enable the Auto-resume feature.
- Solution: Enable "Auto-resume" feature or use "Fill Missing"
- Prevention: N/A (server-side issue)
Google Translate Specific
Google Translate Error
An error occurred while using Google Translate. Please try again later.
- Solution: Wait a few minutes and try again. If the problem persists, try using a different translation model.
- Prevention: N/A (server-side issue)
Network Connection Error
Unable to connect to Google Translate. Please try again later.
- Solution: Wait a few minutes and try again. If the problem persists, try using a different network connection.
- Prevention: N/A (server-side issue)
Unsupported Language
The selected target language is not supported by Google Translate. Please choose a different language.
- Solution: Choose a different target language that is supported by Google Translate.
- Prevention: Check the supported languages list before starting translation.
Invalid Request Format
Invalid translation request format. Please try again.
- Solution: Try refreshing the page and starting the translation again.
- Prevention: Ensure the subtitle content is properly formatted before translation.
Missing Target Language
Target language not properly specified in the translation request.
- Solution: Make sure you have selected a target language before starting translation.
- Prevention: Always select a target language before starting translation.
DeepL Specific
Missing Source Language
Source language not properly specified in the translation request.
- Solution: Make sure you have selected a source language before starting translation.
- Prevention: Always select a source language before starting translation.
Unsupported Source Language
The source language is not supported by DeepL Translate. Please choose a different language.
- Solution: Choose a different source language that is supported by DeepL Translate.
- Prevention: Check the supported languages list before starting translation.
Unsupported Target Language
The target language is not supported by DeepL Translate. Please choose a different language.
- Solution: Choose a different target language that is supported by DeepL Translate.
- Prevention: Check the supported languages list before starting translation.
DeepL XML Parsing Error
DeepL XML parsing error. Please try again.
- Solution: Try use a different model or try again later.
- Prevention: N/A (server-side issue)
Invalid Request Format
Invalid translation request format. Please try again.
- Solution: Try refreshing the page and starting the translation again.
- Prevention: Ensure the subtitle content is properly formatted before translation.
Missing Target Language
Target language not properly specified in the translation request.
- Solution: Make sure you have selected a target language before starting translation.
- Prevention: Always select a target language before starting translation.
Missing API Key
DeepL API key is required. Please provide a valid API key in the settings.
- Solution: Add your DeepL API key in the settings.
- Prevention: Configure your DeepL API key before using DeepL translation.
Authentication Error
DeepL API key is invalid or expired
- Solution: Check and update your DeepL API key in the settings.
- Prevention: Ensure your API key is valid and not expired.
Quota Exceeded
DeepL translation quota exceeded for this billing period, please try using another model or use your own API key.
- Solution: Wait until the next billing period or upgrade your DeepL plan. You can also try using another model or use your own API key.
- Prevention: N/A (server-side issue)
Rate Limit Exceeded
Too many requests to DeepL API. Please try again later
- Solution: Wait a few minutes and try again.
- Prevention: Space out your translation requests.
Connection Error
Unable to connect to DeepL service
- Solution: Check your internet connection and try again.
- Prevention: Ensure stable internet connection before translating.
Temporary Connection Issue
Temporary connection issue with DeepL. Please try again
- Solution: Wait a moment and try again.
- Prevention: N/A (server-side issue)
Glossary Not Found
Specified DeepL glossary not found
- Solution: Check if the glossary exists or create a new one.
- Prevention: Verify glossary availability before use.
General Error
DeepL translation error occurred
- Solution: Try again or switch to a different translation service.
- Prevention: N/A (server-side issue)
Unknown Error
Unknown error occurred while using DeepL service
- Solution: Try again or contact support if the issue persists.
- Prevention: N/A (unexpected error)
I want to translate a video file or a Youtube video, what should I do?
You can use OpenAI's Whisper to transcribe the video, and then use the transcribed subtitle file as the input for the translation. If you just want to transcribe the video file, you can download the tool from this GitHub repository vibe. If you want to download a Youtube video, transcribe it using Whisper and translate it using GPT in one go, I recommend using this tool I've open-sourced on my GitHub repository GPT_subtitles. This repo is the inspiration for this project, and if you find it useful, please consider leaving a star on the repo ⭐. Thank you!
How are the token costs calculated?
The token cost for translations is calculated based on the number of input and output tokens required to process the subtitle content. OpenAI charges different prices for input and output tokens, and the prices vary depending on the model used.
In GPT Subtitler, we use the token consumption of the $0.5/1M Tokens as baseline. Then, we calculate the token consumption based on the actual token usage and the model used, proportionally.
| Model | Input Token Price (per 1M tokens) | Output Token Price (per 1M tokens) | Input Token Multiplier | Output Token Multiplier | Context Window | Max Generation | Output Speed | LMSYS Score |
|---|---|---|---|---|---|---|---|---|
| gemini-3-pro-preview | $2.00 | $12.00 | 4x | 24x | 1M | 16384 | Medium | 1498 |
| gemini-2.5-pro-preview-05-06 | $1.25 | $10.00 | 2.5x | 20x | 1M | 16384 | Slow | 1452 |
| gemini-2.5-pro-preview-06-05 | $1.25 | $10.00 | 2.5x | 20x | 1M | 16384 | Medium | 1452 |
| gemini-2.5-pro-preview-06-05-thinking | $1.25 | $10.00 | 2.5x | 20x | 1M | 16384 | Slow | 1452 |
| gemini-2.5-pro | $1.25 | $10.00 | 2.5x | 20x | 1M | 16384 | Medium | 1452 |
| gemini-2.5-pro-free (Free) | $0.01 | $0.01 | 0.02x | 0.02x | 1M | 16384 | Medium | 1452 |
| gemini-2.5-pro-openai | $1.25 | $10.00 | 2.5x | 20x | 1M | 32768 | Moderately fast | 1452 |
| gemini-2.5-pro-openai-free (Free) | $0.01 | $0.01 | 0.02x | 0.02x | 1M | 8192 | Medium | 1452 |
| chatgpt-4o-latest | $5.00 | $15.00 | 10x | 30x | 128k | 16384 | Moderately fast | 1441 |
| claude-sonnet-4-5-20250929 | $3.00 | $15.00 | 6x | 30x | 200k | 8192 | Moderately fast | 1439 |
| grok-4-latest | $3.00 | $15.00 | 6x | 30x | 256k | 16384 | Moderately fast | 1429 |
| qwen3-235b-a22b-instruct-2507 | $0.29 | $1.14 | 0.572x | 2.286x | 128k | 8192 | Moderately fast | 1428 |
| gpt-5 | $1.25 | $10.00 | 2.5x | 20x | 400k | 128000 | Medium | 1426 |
| gpt-5-chat-latest | $1.25 | $10.00 | 2.5x | 20x | 400k | 128000 | Fast | 1426 |
| gpt-5.1 | $1.25 | $10.00 | 2.5x | 20x | 400k | 128000 | Fast | 1426 |
| openrouter/deepseek/deepseek-r1-0528-free (Free) | $0.01 | $0.01 | 0.02x | 0.02x | 163k | 8192 | Slow | 1424 |
| openrouter/deepseek/deepseek-r1-0528 | $0.50 | $2.18 | 1x | 4.36x | 163k | 8192 | Slow | 1424 |
| grok-4-1-fast-non-reasoning | $0.20 | $0.50 | 0.4x | 1x | 2M | 16384 | Fast | 1420 |
| grok-4-1-fast-reasoning | $0.20 | $0.50 | 0.4x | 1x | 2M | 16384 | Moderately fast | 1420 |
| openrouter/x-ai/grok-4.1-fast (Free) | $0.01 | $0.01 | 0.02x | 0.02x | 2M | 16384 | Fast | 1420 |
| deepseek-chat | $0.28 | $0.42 | 0.56x | 0.84x | 128k | 8192 | Moderately fast | 1415 |
| deepseek-reasoner | $0.28 | $0.42 | 0.56x | 0.84x | 128k | 16384 | Slow | 1415 |
| grok-3-latest | $3.00 | $15.00 | 6x | 30x | 131k | 8192 | Moderately fast | 1409 |
| gemini-flash-latest | $0.30 | $2.50 | 0.6x | 5x | 1M | 16384 | Fast | 1408 |
| gemini-flash-latest-free (Free) | $0.01 | $0.01 | 0.02x | 0.02x | 1M | 16384 | Fast | 1408 |
| gpt-4.1 | $2.00 | $8.00 | 4x | 16x | 1M | 32768 | Moderately fast | 1407 |
| gemini-2.5-flash | $0.30 | $2.50 | 0.6x | 5x | 1M | 16384 | Fast | 1406 |
| gemini-2.5-flash-thinking | $0.30 | $2.50 | 0.6x | 5x | 1M | 16384 | Moderately fast | 1406 |
| gemini-2.5-flash-free (Free) | $0.01 | $0.01 | 0.02x | 0.02x | 1M | 16384 | Fast | 1406 |
| gemini-2.5-flash-thinking-free (Free) | $0.01 | $0.01 | 0.02x | 0.02x | 1M | 16384 | Medium | 1406 |
| claude-haiku-4-5-20251001 | $1.00 | $5.00 | 2x | 10x | 200k | 8192 | Fastest | 1401 |
| qwen3-235b-a22b-thinking-2507 | $0.29 | $2.86 | 0.572x | 5.714x | 128k | 8192 | Medium | 1401 |
| claude-sonnet-4-20250514-thinking | $3.00 | $15.00 | 6x | 30x | 200k | 8192 | Moderately fast | 1399 |
| o4-mini | $1.10 | $4.40 | 2.2x | 8.8x | 200k | 100000 | Moderately fast | 1397 |
| openrouter/deepseek/deepseek-r1 | $0.75 | $2.40 | 1.5x | 4.8x | 64k | 8192 | Slow | 1394 |
| openrouter/deepseek/deepseek-r1-free (Free) | $0.01 | $0.01 | 0.02x | 0.02x | 128k | 128000 | Slow | 1394 |
| siliconflow/deepseek-ai/DeepSeek-R1 | $0.57 | $2.29 | 1.14x | 4.57x | 64k | 8192 | Slow | 1394 |
| gpt-5-mini | $0.25 | $2.00 | 0.5x | 4x | 400k | 128000 | Fast | 1393 |
| claude-sonnet-4-20250514 | $3.00 | $15.00 | 6x | 30x | 200k | 8192 | Moderately fast | 1388 |
| claude-3-7-sonnet-20250219-thinking | $3.00 | $15.00 | 6x | 30x | 200k | 8192 | Moderately fast | 1386 |
| gemini-flash-lite-latest | $0.10 | $0.40 | 0.2x | 0.8x | 1M | 16384 | Fastest | 1381 |
| gemini-flash-lite-latest-free (Free) | $0.01 | $0.01 | 0.02x | 0.02x | 1M | 16384 | Fastest | 1381 |
| qwen3-30b-a3b-instruct-2507 | $0.11 | $0.43 | 0.214x | 0.858x | 128k | 8192 | Fast | 1379 |
| qwen3-30b-a3b-thinking-2507 | $0.11 | $1.07 | 0.214x | 2.142x | 128k | 8192 | Moderately fast | 1379 |
| openrouter/qwen/qwen3-30b-a3b-free (Free) | $0.01 | $0.01 | 0.02x | 0.02x | 41k | 8192 | Fast | 1377 |
| gpt-4.1-mini | $0.40 | $1.60 | 0.8x | 3.2x | 1M | 32768 | Fast | 1372 |
| qwen-max-latest | $2.86 | $8.57 | 5.714x | 17.142x | 32k | 8192 | Medium | 1372 |
| openrouter/qwen/qwen3-235b-a22b-free (Free) | $0.01 | $0.01 | 0.02x | 0.02x | 41k | 8192 | Moderately fast | 1372 |
| claude-3-7-sonnet-20250219 | $3.00 | $15.00 | 6x | 30x | 200k | 8192 | Moderately fast | 1369 |
| gemini-2.0-flash | $0.10 | $0.40 | 0.2x | 0.8x | 1M | 16384 | Fast | 1364 |
| openrouter/google/gemma-3-27b-it-free (Free) | $0.01 | $0.01 | 0.02x | 0.02x | 128k | 8192 | Fast | 1363 |
| grok-3-mini-latest | $0.30 | $0.50 | 0.6x | 1x | 131k | 8192 | Fast | 1359 |
| gemini-2.0-flash-exp-free (Free) | $0.01 | $0.01 | 0.02x | 0.02x | 1M | 16384 | Fast | 1358 |
| gemini-2.0-flash-lite-preview | $0.07 | $0.30 | 0.15x | 0.6x | 1M | 16384 | Fastest | 1351 |
| gemini-2.0-flash-thinking-exp-01-21-free (Free) | $0.01 | $0.01 | 0.02x | 0.02x | 1M | 32768 | Fast | 1351 |
| gemini-2.0-flash-lite-preview-02-05-free (Free) | $0.01 | $0.01 | 0.02x | 0.02x | 1M | 16384 | Fastest | 1351 |
| mistral-small-latest (Free) | $0.01 | $0.01 | 0.02x | 0.02x | 128k | 8192 | Medium | 1348 |
| openrouter/qwen/qwen3-32b-free (Free) | $0.01 | $0.01 | 0.02x | 0.02x | 41k | 8192 | Fast | 1346 |
| o3-mini | $1.10 | $4.40 | 2.2x | 8.8x | 200k | 100000 | Moderately fast | 1344 |
| qwen-plus-latest | $0.11 | $0.29 | 0.228x | 0.572x | 131k | 8192 | Fast | 1344 |
| gpt-4o | $2.50 | $10.00 | 5x | 20x | 128k | 16384 | Moderately fast | 1343 |
| openrouter/qwen/qwq-32b-free (Free) | $0.01 | $0.01 | 0.02x | 0.02x | 128k | 8192 | Fast | 1336 |
| gpt-4o-2024-11-20 | $2.50 | $10.00 | 5x | 20x | 128k | 16384 | Moderately fast | 1318 |
| gpt-4.1-nano | $0.10 | $0.40 | 0.2x | 0.8x | 1M | 32768 | Fastest | 1279 |
| gpt-4o-mini | $0.15 | $0.60 | 0.3x | 1.2x | 128k | 16384 | Fast | 1269 |
| mistral-large-latest (Free) | $0.01 | $0.01 | 0.02x | 0.02x | 128k | 8192 | Medium | 1260 |
| claude-3-5-haiku-20241022 | $1.00 | $5.00 | 2x | 10x | 200k | 8192 | Fastest | 1246 |
| gpt-5-nano | $0.05 | $0.40 | 0.1x | 0.8x | 400k | 128000 | Fastest | 0 |
| open-mistral-nemo (Free) | $0.01 | $0.01 | 0.02x | 0.02x | 128k | 8192 | Fast | 0 |
| pixtral-12b-2409 (Free) | $0.01 | $0.01 | 0.02x | 0.02x | 128k | 8192 | Fast | 0 |
| third-party-model | $0.00 | $0.00 | 1x | 1x | 128k | 8192 | Medium | 0 |
| qwen-turbo-latest | $0.04 | $0.09 | 0.086x | 0.172x | 1M | 8192 | Fastest | 0 |
| qwen-flash | $0.02 | $0.21 | 0.042x | 0.428x | 1M | 8192 | Fastest | 0 |
| google-translate | $0.00 | $0.00 | 0.01x | 0.01x | 100K | 4096 | Fastest | 0 |
| deepl-translate | $2.50 | $2.50 | 5x | 5x | 100K | 4096 | Fastest | 0 |
Please note that while the LMSYS Score represents the general capability of the model, the actual performance for translation tasks may vary for some language pairs. But generally, Gemini models are the best choice for non-English translation.
- The Context Window represents the maximum number of tokens the model can process in a single request.
- Max Generation is the maximum number of tokens the model can generate in a single response.
- Output Speed is a relative measure of how quickly the model generates responses.
- The LMSYS Score is a benchmark score from lmsys.org, indicating the model's general performance across various tasks. (As of 2025-02-06)
The total token cost for a translation request is calculated by multiplying the number of input tokens by the input token price and multiplier, and the number of output tokens by the output token price and multiplier. The sum of these costs gives the final token price for the translation.
You can view the estimated token usage and price for each translation request in the "Translation Cost" section when creating a new translation project. If you have a paid subscription plan, you may have a daily token allowance that you can use for translations without incurring additional costs.
Note: Basic plan users get a 5% discount on token usage, and Premium plan users get a 10% discount on token usage.
Example 1: Using Claude-Haiku model with 1000 input tokens and 2000 output tokens:
Input token = 1000 * 0.5 = 500
Output token = 2000 * 2.5 = 5000
Total token = 500 + 5000 = 5500
Example 2: Using GPT-3.5-TURBO model with 1000 input tokens and 2000 output tokens:
Input token = 1000 * 1.0 = 1000
Output token = 2000 * 3.0 = 6000
Total token = 1000 + 6000 = 7000
Example 3: Premium plan users using Claude-Haiku model with 1000 input tokens and 2000 output tokens:
Input token = 1000 * 0.5 = 500
Output token = 2000 * 2.5 = 5000
Total token = (500 + 5000)*0.9 = 4950
Cache Pricing Information
We use Anthropic's cache service for Claude models and cache the system prompt for each translation request, this can significantly reduce the token usage and cost for subsequent translations if you use the same model and settings and have a long additional context and system prompt. The following table shows the pricing information for cache operations:
For more information about prompt caching, see Anthropic's blog post on prompt caching
| Model | Cache Writes Token Price | Cache Hits Token Price | Cache Writes Token Multiplier | Cache Hits Token Multiplier |
|---|---|---|---|---|
| claude-3-5-haiku-20241022 | $1.25 | $0.10 | 2.5x | 0.2x |
| claude-haiku-4-5-20251001 | $1.25 | $0.10 | 2.5x | 0.2x |
| claude-3-7-sonnet-20250219 | $3.75 | $0.30 | 7.5x | 0.6x |
| claude-3-7-sonnet-20250219-thinking | $3.75 | $0.30 | 7.5x | 0.6x |
| claude-sonnet-4-20250514 | $3.75 | $0.30 | 7.5x | 0.6x |
| claude-sonnet-4-20250514-thinking | $3.75 | $0.30 | 7.5x | 0.6x |
| claude-sonnet-4-5-20250929 | $3.75 | $0.30 | 7.5x | 0.6x |
How do transcription models compare in terms of cost and accuracy?
Our platform offers several transcription models with different cost structures and performance characteristics. The table below compares these models based on token cost per minute, processing speed, word error rate, and language support.
| Model | Tokens/Minute | Speed | Word Error Rate | Languages |
|---|---|---|---|---|
| AssemblyAI Slam-1 (English Only) | 12,000 | Medium | 8.6% | English only |
| AssemblyAI Universal | 12,000 | Medium | 8.7% | 102 languages |
| Whisper Large v3 | 8,000 | Medium | 10.3% | 99 languages |
| Wizper | 4,000 | Fast | 10.3% | 99 languages |
| Mistral Voxtral Mini | 2,000 | Fast | 11.2% | Multilingual |
| AssemblyAI Nano | 5,000 | Fast | 12.7% | 102 languages |
Important Notes on Transcription Models
- Token cost is calculated based on the duration of the audio file, with different models having different token rates per minute.
- Word Error Rate (WER) is a measure of transcription accuracy - lower percentages indicate better performance.
- Processing speed indicates how quickly the model can transcribe audio relative to its duration.
- Number of languages supported by each model. Some models are optimized for specific languages.
- AssemblyAI models are billed per second with rounding to the nearest second.
- FAL AI models (Whisper, Wizper) are billed by exact duration with no rounding.
For general use, we recommend Whisper Large v3 for the best balance of accuracy and cost. For English-only content requiring highest accuracy, AssemblyAI Slam-1 is recommended. For cost-sensitive applications, AssemblyAI Nano offers good performance at a lower price point.
Can I use my own API key for translations?
Yes, you can choose to use your own OpenAI/Claude/Gemini API key for translation. If you provide an API key, the server will use it for translation, and your account Token balance and usage will not be affected. If you use a paid subscription plan, you can also use the translation service without providing an API key and without using Token balance, until your Token quota is exhausted.
How do I get an OpenAI/Gemini API key?
To get an OpenAI API key, you need to register an account on the OpenAI website at openai signup and openai api-keys . Once you have the API key, you can use it in the translation settings to enable the use of the OpenAI API for translations. This website will not store your API key or use it in any other way than for the translation process. It is recommended to revoke your API key once in a while to ensure the security of your account.
To get a free Gemini API key, visit https://aistudio.google.com/app/apikey. Follow the instructions to generate your API key. Once you have the API key, you can use it in the translation settings to enable the use of the Gemini API for translations.
How can I check my token balance and usage?
- Sign in to your account.
- Navigate to the "Settings" page.
- On the settings page, you will find information about your token balance, subscription tier, and usage details.
- You can also view your token usage history and transaction logs on the same page.
What is the Whisper transcription feature?
The Whisper transcription feature is a new addition to our service that allows you to automatically transcribe audio and video files. It uses OpenAI's Whisper model to generate accurate transcriptions in multiple languages. This feature is particularly useful for creating subtitles for videos, transcribing interviews, or converting speech to text for various purposes. You can access this feature in the 'Audio Transcriber' section of our website.
How do I use the Whisper transcription feature?
- Go to the 'Audio Transcriber' section of the website. Try it now!
- Upload your audio or video file, or provide a YouTube URL.
- Select the language of the audio content (if known).
- Optionally, add any specific instructions or context in the 'Prompt' field.
- Click the 'Start Transcription' button to start the process.
- Once completed, you can view, edit, and download the transcription, or create a new subtitle project with the result.
How does batch translation work?
The batch translation feature allows you to translate multiple subtitle files in one project. Here are the key points:
File Limits:
- Maximum 10 files per batch project
- Free users: translate up to 2 files simultaneously
- Basic/Premium users: translate up to 5 files simultaneously
Translation Process:
- Create a new batch project
- Upload your SRT files (drag & drop supported)
- Select files to translate
- Configure translation settings once for all files
- Monitor progress and download completed translations
Tips:
- Avoid using free model to translate too many files at once as it may cause rate limit issue
- Preview files before translation
- Download individual files or all completed translations at once
- Stop and resume translations as needed
What languages are supported for translation?
GPT Subtitler supports translation between all languages supported by the language models, including English, Chinese, Spanish, Japanese, Korean, and more. It can even translate some fictional languages, such as Elvish or Valyrian. When creating a new translation project, you can select the source and target languages from the provided dropdown menus. If your desired language is not listed, you can select "Other" and manually enter the language. Please note that the quality of translation may vary depending on the language pair and the model used. Currently, this project is optimized for English to Chinese translation, other language pairs may have different effects.
How are the translations processed?
The backbone of the translation service of our website is available in GitHub GPT_subtitles , but in short, the translations are processed by sending the subtitle content in batches to the OpenAI/Claude/Gemini API for processing. The prompt is carefully crafted to guide the model to generate accurate translations. We also utilize additional context provided by the user and few-shot examples to improve the translation quality. The API generates the translated text based on the input content and the specified settings, such as source and target languages, prompt, and model. The translated text is then displayed in real-time on the website, allowing you to monitor the progress and download the translated subtitles once the process is complete.
What are few-shot examples and how do they help with translations?
Few-shot examples are a small set of example translations provided to the AI model to guide and improve the quality of the generated translations. By including a few relevant examples of how specific phrases or sentences should be translated from the source language to the target language, the model can better understand the desired translation style, tone, and context. This approach helps to produce more accurate and contextually appropriate translations, especially for domain-specific or idiomatic expressions. The minimal required format for few-shot examples is as follows:
[\
{\
"input": {\
"current_batch_subtitles": [\
{\
"index": 1,\
"original_text": "This is a sample subtitle."\
}\
]\
},\
"output": {\
"current_batch_subtitles_translation": [\
{\
"index": 1,\
"original_text": "This is a sample subtitle.",\
"first_translation": "这是一个示例字幕。",\
"translation": "这是一个示例字幕。"\
}\
]\
}\
}\
]
Optionally, you can include 'previous_batch_subtitles' and 'next_batch_subtitles' fields to provide additional context for the translations. These fields should be placed inside the 'input' object, alongside the 'current_batch_subtitles' field. The structure for these optional fields is as follows:
{
"previous_batch_subtitles": [\
{\
"index": 1,\
"original_text": "Previous subtitle 1",\
"first_translation": "之前的字幕1",\
"translation": "之前的字幕1"\
}\
],
"next_batch_subtitles": [\
{\
"index": 3,\
"original_text": "Next subtitle 1"\
}\
]
}
Prompt for generating few-shot examples. After generating the examples, copy the final JSON array into the few-shot examples section. This should improve the accuracy of the translation and reduce the error rate of the model:
Source Language: (fill in here)
Target Language: (fill in here)
I need few-shot examples for translating subtitles from [Source Language] to [Target Language]. Please provide 5-7 examples that showcase common subtitle structures, idiomatic expressions, and cultural references. Each example should include the original text in [Source Language], its translation in [Target Language], and any relevant context or explanation. The examples should be formatted as follows:
{
"input": {
"current_batch_subtitles": [\
{\
"index": 1,\
"original_text": "[Source Language Text]"\
}\
]
},
"output": {
"current_batch_subtitles_translation": [\
{\
"index": 1,\
"original_text": "[Source Language Text]",\
"translation": "[Target Language Translation]"\
}\
]
}
}
Please ensure the examples cover a range of scenarios typical in subtitle translation.
After providing the individual examples, please combine all the examples into a single JSON array structure, like this:
[\
{\
"input": { ... },\
"output": { ... }\
},\
{\
"input": { ... },\
"output": { ... }\
},\
...\
]
This combined JSON structure should include all the examples you've created, making it easy to use as a complete set of few-shot examples for the translation model.
Copy Prompt
What is the best model for language translation accuracy? Are there any settings you can suggest?
Based on experience, Gemini-pro generally offers the best performance for translation tasks. However, the optimal model can vary depending on the specific language pair. If Gemini-pro's performance doesn't meet your expectations, try the following models in this order:
1. Gemini-pro
2. Claude-3.5-sonnet
3. GPT-4o
4. GPT-4o-mini
5. Gemini-flash
6. Claude-haiku
It's recommended to test these models with a smaller batch of text to determine which works best for your specific use case. Note that DeepSeek models are exceptionally good and cost-effective for Chinese translation but may not perform as well for other languages.
Recommended settings:
1. Batch Size: Use a higher batch size to provide more context for the translation.
2. Enable Context Memory: This helps the model better understand the video context and produce more accurate translations, although it may slightly increase token usage.
3. Fluent Merging: Beneficial when using the Whisper model to generate subtitles. It helps handle cases where sentences are broken across multiple lines, creating more natural translations.
4. Few-shot examples: These can significantly improve translation accuracy by providing guidance to the model. As these are language-specific, you may need to create custom examples for your language pair.
A setting incorporating many of these features is available at: https://gptsubtitler.com/settings/8dd9a960-cb30-4141-8b23-d3944d97773e. You can access and modify it to fit your specific needs.
How do I share my translation settings?
In the settings modal, click on the 'Share' button next to your setting. You can then review and edit your setting, after publish the setting, you can use "copy link" to get a shareable link.
Can I use settings created by other users?
Yes! You can access the translation settings by opening link shared by others, bookmark ones you like, and load them into your translation settings.
How do I change my username?
Go to your account settings. You'll find an option to edit your username next to your username. If you haven't set one, a random username will be assigned which you can then change.
How do I write effective custom instructions?
Custom instructions in the Additional Context field give the AI model specific guidance on how to translate your subtitles. Here are some tips for writing effective instructions:
1. Be specific and clear: Instead of "make it sound good", try "translate using casual language appropriate for a teenage audience".
2. Focus on one aspect at a time: Separate formatting instructions ("preserve line breaks") from style instructions ("use formal language").
3. Provide examples when possible: If you want specific terms translated in a certain way, include examples.
4. Consider cultural context: If the content has cultural references, specify how to handle them ("explain Japanese cultural references briefly in parentheses").
5. Use the Instruction Library: Access pre-made instruction templates by clicking the "Instruction Library" button below the Additional Context field.
Remember that instructions should be in the target language or English, as they're meant to guide the AI, not to be translated themselves.
How can I contact you?
If you have any further questions, feedback, or need assistance, please feel free to contact me at support@gptsubtitler.com . I will be happy to help you with any queries or issues you may have.
How can I get a refund?
I'm sorry that we didn't meet your expectations. We support refunds within 7 days after payment. Please contact me through the email you used when registering ( support@gptsubtitler.com), and I will do my best to resolve your issue.
Ready to Transform Your Subtitles?
Join thousands of content creators who are already using GPT Subtitler to reach global audiences.
Try NowView Pricing
StripeM-Inner