API Translation
mobile
pc
cpu
api
ai
モデルDLなし / WebGPU不要 / 自分の翻訳API。
TransPackgenerator
WebGPU + WebLLM Edition
Menu
Upload Fabric/Forge/NeoForge .jar/.zip and build a translated Minecraft resource pack fully in-browser.
Step 1: Select .jar / .zip / .json
Drag and drop Fabric/Forge/NeoForge mod(.jar), shader pack(.zip), or lang json(.json)
Step 2: Translation target language
Step 3: Translation settings
For low-end PCs, 30 lines / 1 job is recommended.
Step 4: Model selection
API settings
Saved automatically in this browser.
Model download progress
Not started
Translation progress
0 / 0 lines
Task status
Waiting...
進捗情報の詳細
まだ詳細ログはありません。
Model guide (recommended specs)
Gemma-2-9B
High accuracy
pc
gpu
NVIDIA
Radeon
Intel
約DL: 1.9GB / 推奨VRAM: 4〜6GB / RAM: 16GB以上。
Gemma-2-2B
Fast
Lightweight
mobile
pc
gpu
NVIDIA
Radeon
Intel
約DL: 1.9GB / 推奨VRAM: 4〜6GB / RAM: 16GB以上。
Gemma-4-E2B
Accuracy focused
pc
gpu
NVIDIA
Radeon
Intel
約DL: 7.2GB / 推奨VRAM: 10GB以上 / RAM: 32GB以上。
Qwen3.5-9B
High accuracy
pc
gpu
NVIDIA
Radeon
Intel
約DL: 4.3GB / 推奨VRAM: 7〜10GB / RAM: 24GB以上。
Llama-3.1-8B-Instruct
High accuracy
pc
gpu
NVIDIA
Radeon
Intel
約DL: 5.0GB / 推奨VRAM: 8〜10GB / RAM: 24GB以上。
Llama-3.2-3B-Instruct
Fastest class
mobile
pc
gpu
NVIDIA
Radeon
Intel
約DL: 1.8GB / 推奨VRAM: 3〜5GB / RAM: 16GB以上。
Qwen3-4b-2507
Balanced
mobile
pc
gpu
NVIDIA
Radeon
Intel
約DL: 2.5GB / 推奨VRAM: 4〜7GB / RAM: 16GB以上。
Qwen-2.5-7B-Instruct
High accuracy
pc
gpu
NVIDIA
Radeon
Intel
約DL: 4.3GB / 推奨VRAM: 7〜10GB / RAM: 24GB以上。
Hermes-3-Llama-3.1-8B
High accuracy
pc
gpu
NVIDIA
Radeon
Intel
約DL: 4.9GB / 推奨VRAM: 8〜10GB / RAM: 24GB以上。
タグ説明: mobile = 軽めの端末向け、pc = PC向け、cpu = CPU/APIルート、gpu = ローカルWebGPUルート、api = 外部翻訳APIルート。
Build Minecraft translation packs without unpacking files by hand
TransPackgenerator reads language files from Minecraft mod archives, shader packs, or standalone lang JSON files, translates only the value strings, and returns a resource pack ZIP that can be loaded by Minecraft.
What it preserves
Translation keys, formatting codes, numbered placeholders, escaped newlines, and common Minecraft tokens are protected during translation so the generated pack remains usable in game.
Supported inputs
Use Fabric, Forge, NeoForge, or compatible mod archives, shader pack ZIP files, and individual language JSON files. The tool searches for language assets and builds the matching resource pack structure.
Translation modes
CPU/API mode works on mobile and GPU-less PCs without a local model download. WebGPU mode runs supported local WebLLM models in the browser for users with compatible hardware.
Recommended workflow
For the most reliable result, start with a small language file or a single mod, check the generated pack in Minecraft, then process larger mod packs once the language and model settings are correct.
1. Choose the source file
Select a mod archive, shader pack, or lang JSON. The file name is shown before translation starts.
2. Pick the target language
Search the language list and choose one Minecraft locale, or choose All languages to write every supported locale into the generated pack.
3. Select CPU/API or WebGPU
Use CPU/API mode for broad compatibility, or a local model when your browser and GPU support WebGPU.
4. Review the output
Install the downloaded resource pack, open Minecraft, and check important screens for context-sensitive wording.
Privacy, availability, and responsible use
The site is designed as a public utility page with clear navigation, permanent contact links, and crawlable policy pages. No login is required to use the generator.
File handling
Archive parsing and ZIP creation happen in the browser. When CPU/API translation is selected, extracted text values are sent to this site's translation endpoint and may be forwarded to the configured translation providers.
Responsible localization
Generated translations should be reviewed before publication. Respect mod licenses, resource pack rules, and the original authors' distribution terms.
More information
Read the help, privacy, terms, license, and contact pages for the full workflow, data handling notes, and support channels.