Skip to main content
PromptQuorum
Home/Power Local LLM/TranslateBooksWithLLMs Review: Translate Full Books With a Local or Cloud LLM
Productivity & Knowledge Tools

TranslateBooksWithLLMs Review: Translate Full Books With a Local or Cloud LLM

Β·11 min readΒ·By Hans Kuepper Β· Founder of PromptQuorum, multi-model AI dispatch tool Β· PromptQuorum
Read in:
πŸ‡ΊπŸ‡ΈenπŸ‡©πŸ‡ͺdeπŸ‡«πŸ‡·frπŸ‡―πŸ‡΅jaπŸ‡¨πŸ‡³zhπŸ‡ͺπŸ‡ΈesπŸ‡§πŸ‡·ptπŸ‡ΈπŸ‡¦arπŸ‡°πŸ‡·ko

TranslateBooksWithLLMs (TBL) is a free, open-source application that translates full-length books, subtitles, and documents β€” EPUB, DOCX, TXT, and SRT β€” using either a locally hosted model through Ollama or a cloud LLM provider such as OpenAI, Gemini, Mistral, DeepSeek, Poe, or OpenRouter. It ships as a packaged desktop app for Windows and macOS (Intel and Apple Silicon), a self-hosted web interface, and a command-line tool, with no file-size limit thanks to an intelligent chunking system that preserves formatting and can resume an interrupted translation from a checkpoint. It is licensed under the AGPL-3.0, a copyleft license, and has over 2,400 GitHub stars as of this review.

TranslateBooksWithLLMs (github.com/hydropix/TranslateBooksWithLLMs), often shortened to "TBL," is a free, open-source application for translating full-length books, subtitles, and documents using either a local model through Ollama or a cloud LLM provider. It ships as a downloadable desktop app for Windows and macOS, a browser-based web UI (localhost:5000) for anyone running it from source or Docker, and a command-line interface for scripted or batch use. This review covers what it actually does, how to install it, what it costs, and who it fits. This review is a companion piece to TranslateBooksWithLLMs's entry in the Local LLM Software Directory.

Key Takeaways

  • TranslateBooksWithLLMs (github.com/hydropix/TranslateBooksWithLLMs), or "TBL," is a free, open-source app for translating full books, subtitles, and documents
  • Built by a developer using the GitHub handle hydropix; this review found no company, funding round, or commercial backer behind the project
  • AGPL-3.0 licensed, confirmed via the GitHub repository's license metadata
  • Supports EPUB, DOCX, TXT, and SRT formats, with no file-size limit via intelligent, token-based chunking that preserves formatting and subtitle timecodes
  • Works with a local Ollama model for free, or cloud providers β€” OpenAI, Google Gemini, Mistral, DeepSeek, Poe, OpenRouter, NVIDIA NIM β€” plus any OpenAI-compatible local server such as llama.cpp, LM Studio, or vLLM
  • Ships a checkpoint/resume system, optional glossary and writing-style extraction for consistent terminology and tone across a whole book, and optional text-to-speech output via Edge-TTS
  • Over 2,400 GitHub stars and 319 forks as of this review; the repository was created May 22, 2025

πŸ“ In One Sentence

TranslateBooksWithLLMs (TBL) is a free, open-source (AGPL-3.0) application, available as a Windows/macOS desktop app, a web UI, and a CLI, that translates full-length books, subtitles, and documents (EPUB, DOCX, TXT, SRT) using either a local Ollama model or a cloud LLM provider, with no file-size limit and a checkpoint system that resumes interrupted jobs.

πŸ’¬ In Plain Terms

Instead of pasting a book chapter by chapter into a chat window, TranslateBooksWithLLMs takes the whole file, breaks it into manageable pieces behind the scenes, translates each piece while keeping the surrounding context and formatting intact, and reassembles the finished file β€” resuming automatically if the process gets interrupted partway through.

πŸ“ŒNote: This review is based on TranslateBooksWithLLMs's own GitHub repository, README, and documentation, not on independent PromptQuorum translation-quality testing. The project links its own community-maintained translation-quality benchmarks on its GitHub wiki β€” treat those as self-published, first-party data rather than an independently audited benchmark.

What Is TranslateBooksWithLLMs?

TranslateBooksWithLLMs is a free, open-source application built around one task: translating a complete book, document, or subtitle file with an LLM, rather than translating a page or a chat message at a time. Its own GitHub description states it can "Translate full-length books and documents with Ollama, OpenAI-compatible, Gemini, Mistral, DeepSeek, Poe or OpenRouter," and that it "Preserves formatting. Resumes where you left off. No file size limits."

  • Product type: a self-contained translation application with three interfaces β€” a packaged desktop app, a local web UI, and a command-line tool β€” not a plugin for another program
  • Creator: a developer using the GitHub handle hydropix; no legal or company name is published in the repository as of this review
  • License: AGPL-3.0, a copyleft open-source license, confirmed via the GitHub repository's license metadata
  • Repository: github.com/hydropix/TranslateBooksWithLLMs, created May 22, 2025
  • Scale: over 2,400 GitHub stars, 319 forks, 11 contributors, and 25 open issues as of this review
  • Core engine: a Python translation pipeline that chunks input text, sends each chunk to whichever LLM provider you configure, and reassembles the output while preserving structure

Project History and Version Milestones

TranslateBooksWithLLMs's GitHub repository was created on May 22, 2025, and has shipped continuously since, moving from a command-line-first translation script to a full desktop application with a shared, multi-device web workspace. Version numbers below come from the project's own GitHub releases; check the releases page directly for anything shipped after this review's publish date.

  1. 1
    Repository created β€” May 22, 2025
    Why it matters: The project begins as a Python-based book and document translation tool built around Ollama and other LLM providers.
  2. 2
    v1.4.x series β€” May–August 2026
    Why it matters: A long run of minor releases (v1.4.1 through v1.4.11) through mid-2026, per the GitHub releases list, indicating steady incremental feature and fix delivery.
  3. 3
    v1.5.0 β€” August 2, 2026
    Why it matters: A new minor version line begins, per the GitHub releases list; check the specific release notes on GitHub for what changed.
  4. 4
    v1.5.6–v1.5.9 β€” August 9–24, 2026
    Why it matters: Continued patch-level iteration through August 2026, per the GitHub releases list.
  5. 5
    v1.5.10 β€” September 6, 2026: most recent tagged release as of this review
    Why it matters: The current release at the time of this review; the desktop app, checkpoint/resume system, glossary and style extraction, and multi-provider support described in this review are all present by this version.

What Can You Do With TranslateBooksWithLLMs?

TranslateBooksWithLLMs's feature set centers on getting a complete, correctly formatted translated file out the other end of an LLM, for documents far longer than a single chat context window. Here is what each part actually does, per the project's own README and documentation.

  • Unlimited-length translation via chunking β€” the app splits input into token-based chunks (450 tokens by default, per the documented .env configuration) and translates each one while preserving context between segments, so book-length input is not a hard limit
  • Format preservation β€” EPUB structure and styling, DOCX formatting, and SRT subtitle timecodes are kept intact in the translated output, per the project's own description
  • Checkpoint and resume β€” an interrupted translation (closed app, lost connection, rate limit) can be resumed from where it left off rather than restarted from the beginning
  • Reusable writing-style presets β€” extract a style preset from sample books (or write one by hand) and apply it to every chunk of a new translation for a consistent register and tone across the whole book
  • Auto glossary and style extraction β€” if you have no preset ready, selecting "Auto" derives both a glossary and a style profile directly from the document being translated, using one additional LLM call before the job starts
  • Multi-provider support β€” a local Ollama model for free, any OpenAI-compatible local server (llama.cpp, LM Studio, vLLM, LocalAI), or a cloud provider: OpenAI, Google Gemini, Mistral, DeepSeek, Poe, OpenRouter, or NVIDIA NIM
  • API key rotation β€” configure a comma-separated list of API keys for a single provider, and the app automatically rotates to the next key when one hits a rate limit
  • Multi-device shared web workspace β€” the web UI has no user accounts; every browser pointed at the same running server sees the same job progress, history, and most preferences in near real time
  • Optional text-to-speech output β€” generate audio from the translated text via Edge-TTS, alongside the translated document itself
  • Notifications β€” get a push notification (via ntfy, Discord, Slack, gotify, or a custom HTTP webhook) when a long translation job finishes

Usage Examples: Three Ways to Use TranslateBooksWithLLMs

These are concrete workflows built from TranslateBooksWithLLMs's own documented features above, not hypothetical use cases.

TranslateBooksWithLLMs Pricing and Licensing

TranslateBooksWithLLMs itself is free and open source, licensed under the AGPL-3.0. There is no paid tier, subscription, or account requirement for the application. Your actual cost depends entirely on which model provider you translate against.

  • The application is free, with no usage limits, watermarks, or feature gates imposed by the project itself
  • Translating with a local Ollama model costs only your own hardware and electricity β€” no per-word or per-token fee to TranslateBooksWithLLMs or to Ollama
  • Translating with a cloud provider (OpenAI, Gemini, Mistral, DeepSeek, Poe, OpenRouter, or NVIDIA NIM) costs whatever that provider charges for API usage; several of these providers publish a free tier, but TranslateBooksWithLLMs does not control or guarantee that pricing
  • License: AGPL-3.0, confirmed via the GitHub repository's license metadata β€” this is a copyleft license with a network-use clause, meaning that if you modify the code and run it as a network service for others, you are generally required to make your modified source available too; this differs meaningfully from a permissive license like MIT or Apache 2.0, so review the license text yourself if that obligation matters for your use case
  • This review found no evidence of a funding round, company, or commercial backer behind the project β€” its own README links a Ko-fi page for voluntary donations, consistent with an independently maintained, community-supported open-source project

Local (Ollama) vs. Cloud Translation in TBL

TranslateBooksWithLLMs supports both a fully local translation path through Ollama and a cloud-provider path, and the two differ mainly in speed, cost, and privacy β€” not in which file formats or features are available.

Cost

Local (Ollama):
Free beyond your own hardware/electricity
Cloud provider:
Billed by the provider per API usage; several offer a free tier

Privacy

Local (Ollama):
Text is not sent anywhere for translation β€” it stays on your machine
Cloud provider:
Text is sent to the provider's servers per their own data-handling terms

Concurrency

Local (Ollama):
Serialized β€” one chunk at a time (PARALLEL_TRANSLATIONS is ignored)
Cloud provider:
Supports --parallel N for concurrent chunk translation, subject to the provider's own rate limits

Setup

Local (Ollama):
Requires installing Ollama and pulling a model locally first
Cloud provider:
Requires only an API key from the chosen provider

Translation quality

Local (Ollama):
Depends on the local model's size and language coverage
Cloud provider:
Depends on the chosen cloud model; typically a wider choice of larger models

The project links its own community translation-quality benchmarks, organized by target language, on its GitHub wiki β€” check that page directly to compare specific local and cloud models for your target language rather than relying on general assumptions about local-vs-cloud quality.

Who Should Use TranslateBooksWithLLMs?

TranslateBooksWithLLMs fits people who need a whole book, document, or subtitle file translated end-to-end with consistent formatting and terminology, rather than a short snippet pasted into a chat window.

Competitors and Alternatives

Dedicated, document-length LLM book-translation tools are a narrow niche on PromptQuorum's current review catalog β€” TranslateBooksWithLLMs does not have a close, feature-matched rival among the site's existing FeatureAppPost reviews. The closest available comparisons are other local-LLM-powered document and knowledge tools that share TranslateBooksWithLLMs's emphasis on processing whole documents locally, even though neither is a translation-specific tool.

Karakeep

Best known for:
Self-hostable bookmark-and-notes app with AI-based tagging, summarization, and search over saved documents
Link:
Karakeep review
Articles about Karakeep (2)

Also mentioned in:

Khoj

Best known for:
Self-hostable, open-source personal AI "second brain" answering questions over your own documents
Link:
Khoj review
Articles about Khoj (2)

Also mentioned in:

Neither Karakeep nor Khoj translates documents β€” they are included here as the closest local-LLM-powered, document-oriented productivity tools already reviewed on PromptQuorum, not as direct feature-for-feature TranslateBooksWithLLMs alternatives. See the Local LLM Software Directory for the full catalog if a closer match is added later.

Common Mistakes When Evaluating TranslateBooksWithLLMs

Most confusion about TranslateBooksWithLLMs comes from its licensing terms, assuming it works like a live chat translator, or missing that local translation runs one chunk at a time.

Frequently Asked Questions

What is TranslateBooksWithLLMs?

TranslateBooksWithLLMs (github.com/hydropix/TranslateBooksWithLLMs), or "TBL," is a free, open-source (AGPL-3.0) application for translating full-length books, subtitles (SRT), and documents (EPUB, DOCX, TXT) using either a local Ollama model or a cloud LLM provider.

Is TranslateBooksWithLLMs free?

Yes, the application itself is free and open source with no paid tier. If you translate with a local Ollama model, the only cost is your own hardware and electricity. If you connect a cloud provider (OpenAI, Gemini, Mistral, DeepSeek, Poe, OpenRouter, or NVIDIA NIM), you pay that provider directly for API usage.

How do I install TranslateBooksWithLLMs?

Download the packaged desktop app for Windows or macOS (Intel or Apple Silicon) from the latest GitHub release, extract it, and run the executable β€” the web UI opens at localhost:5000. Linux users run it from source (git clone plus ./start.sh) or via Docker, since there is no prebuilt Linux binary as of this review.

Does TranslateBooksWithLLMs work offline?

Yes, if you use a local model through Ollama. Once Ollama and a model are installed, the translation itself runs entirely on your machine β€” the app does not send book or document text to an external server for translation. (It may still make ordinary background network calls, such as checking for app updates, unrelated to your document content.) Using a cloud provider instead requires an internet connection and sends your text to that provider.

What file formats does TranslateBooksWithLLMs support?

EPUB, DOCX, TXT, and SRT (subtitles), per the project's own documentation. It preserves EPUB formatting and structure, DOCX styling, and SRT timecodes in the translated output.

Is there a file-size limit?

No. TranslateBooksWithLLMs uses a token-based chunking system (450 tokens per chunk by default) that processes documents of any length, including thousand-page books, while preserving context between chunks.

Can I resume an interrupted translation?

Yes. TranslateBooksWithLLMs's checkpoint system saves progress automatically, so an interrupted job β€” from a closed app, a lost connection, or a rate limit β€” can resume from where it left off rather than restarting from the beginning.

Which cloud providers does TranslateBooksWithLLMs support?

OpenAI, Google Gemini, Mistral, DeepSeek, Poe, OpenRouter, and NVIDIA NIM, plus any OpenAI-compatible local server such as llama.cpp, LM Studio, or vLLM, per the project's own README.

What license does TranslateBooksWithLLMs use?

AGPL-3.0, a copyleft open-source license with a network-use clause. If you modify the code and run it as a hosted service for other users, you are generally required to release your modified source too β€” read the license text directly if that applies to your planned use.

Who created TranslateBooksWithLLMs?

A developer using the GitHub handle hydropix, per the repository's ownership. This review found no published legal or company name, and no evidence of a funding round behind the project.

Sources

← Back to Power Local LLM