Owl-L Quality mode packs

Byte-identical mirrors of open-source files that Owl-L (a free, offline translator) downloads when you turn on Quality mode. Nothing was converted, re-quantized or modified: the sha256 of every file equals that of the original, and Owl-L checks it after every download. They are mirrored here only so that Owl-L does not depend on third-party repositories staying put. The original authors made the files; all credit belongs to them.

Every file keeps its original license; each license text is in licenses/. The licenses allow redistribution under their conditions (copyright and license notice kept, which is what this repository does). Each license applies only to its own file, not to the repository as a whole.

File What it is Original source License Size (bytes) sha256
Hy-MT2-1.8B-Q4_K_M.gguf Hy-MT2 1.8B translation model, Q4_K_M, by Tencent tencent/Hy-MT2-1.8B-GGUF @ a0c709d9fac510f2c807aa3af52872340dc37a4a Apache-2.0, licenses/Hy-MT2-1.8B-LICENSE.txt 1133080448 dc5f44fcf1fa496ee7ad725982c0c8c553a4de00259b53af84c4b89fb0c06699
Index-Translate-2B.Q4_K_M.gguf Index-Translate 2B translation model, Q4_K_M, by IndexTeam (bilibili) IndexTeam/Index-Translate-2B-GGUF @ 3a0eca6ce931d88d00614d427ab3ed8ff0743902; code and license: bilibili/Index-Translate Apache-2.0, licenses/Index-Translate-LICENSE.txt 1312164352 044b313d29342bd3b2c77cbb64023ca0d209bd9b3247763f9d162767ef2d746a
Phi-4-mini-instruct-Q4_K_M.gguf Phi-4-mini-instruct, Q4_K_M GGUF by Unsloth of the model by Microsoft unsloth/Phi-4-mini-instruct-GGUF @ 78eb92a46fc37e6b524df991ed9aca9bc6aa7b80; original: microsoft/Phi-4-mini-instruct MIT, licenses/Phi-4-mini-LICENSE.txt, licenses/Phi-4-mini-NOTICE.md 2491874272 88c00229914083cd112853aab84ed51b87bdf6b9ce42f532d8c85c7c63b1730a
llama-b12018-bin-win-cpu-x64.zip llama-server and the llama.cpp tools, build b12018, Windows x64 CPU, by the ggml-org / llama.cpp contributors ggml-org/llama.cpp release b12018 MIT, licenses/llama.cpp-LICENSE.txt (the zip also contains LICENSE-LLVM-OpenMP) 24574711 205755a92c77f187886a77854c8a7fd018ad8bbe17e4a8d128a032f1aaacb9a6

The Hy-MT2 license text is the one published in the original repository: "Hy-MT2-1.8B-GGUF is licensed under the Apache License, Version 2.0" (Copyright 2026 Tencent). The IndexTeam Hugging Face repositories carry the Apache-2.0 tag in their metadata and no license file; the license text here is the root LICENSE of the authors' GitHub repository, whose README states "Apache-2.0" for the project, at the commit named in SOURCES.md.

Not affiliated with Tencent, IndexTeam, bilibili, Microsoft, Unsloth or ggml-org. If you are an author and want a file removed or the credit changed, open a discussion here.

Downloads last month
72
GGUF
Model size
2B params
Architecture
hunyuan-dense
Hardware compatibility
Log In to add your hardware

4-bit

Inference Providers NEW
This model isn't deployed by any Inference Provider. 🙋 Ask for provider support