nsugar commited on
Commit
10b0993
·
verified ·
1 Parent(s): a46fea2

emu 1.0.0: THIRD_PARTY_LICENSES.md

Browse files
Files changed (1) hide show
  1. THIRD_PARTY_LICENSES.md +79 -0
THIRD_PARTY_LICENSES.md ADDED
@@ -0,0 +1,79 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ # Third-Party Licenses & Attributions
2
+
3
+ emu is a proprietary product of CloudKites Pty Ltd (see `LICENSE`). It bundles or
4
+ downloads the third-party AI models and software components listed below, each
5
+ used under its own license. The notices in this file are provided to comply with
6
+ those licenses and are part of the emu End-User License Agreement.
7
+
8
+ emu does **not** modify the third-party model weights' architectures; the models
9
+ are run for inference and, where noted, quantized for on-device efficiency. Where
10
+ a license requires that modifications be indicated (e.g. CC-BY-4.0), quantization
11
+ and format conversion to emu's bundle format are the only changes made.
12
+
13
+ ---
14
+
15
+ ## AI models
16
+
17
+ | Model | Role in emu | Provider | License |
18
+ |---|---|---|---|
19
+ | **Gemma 4 12B** (instruction-tuned) | Chat (text) | Google DeepMind | Gemma Terms of Use |
20
+ | **Gemma 4 E4B** (instruction-tuned) | Chat + multimodal (text · image · audio) | Google DeepMind | Gemma Terms of Use |
21
+ | **Qwen 3.5 4B** | Chat (text) | Alibaba Cloud (Qwen) | Apache License 2.0 |
22
+ | **Qwen 3.5 9B** | Chat (text) | Alibaba Cloud (Qwen) | Apache License 2.0 |
23
+ | **Qwen3-ASR 0.6B** | Speech-to-text | Alibaba Cloud (Qwen) | Apache License 2.0 |
24
+ | **Qwen3-TTS 12Hz 0.6B** | Text-to-speech (multilingual) | Alibaba Cloud (Qwen) | Apache License 2.0 |
25
+ | **gwen-tts 0.6B** (Vietnamese voice cloning; fine-tune of Qwen3-TTS) | Text-to-speech | G-Group AI Lab | MIT License |
26
+ | **Pocket TTS** | Text-to-speech (fast English) | Kyutai | CC-BY-4.0 |
27
+ | **VieNeu-TTS v3-Turbo** (built on MOSS-TTS-Nano + MOSS-Audio-Tokenizer) | Text-to-speech (Vietnamese · English) | pnnbao-ump; OpenMOSS Team | Apache License 2.0 |
28
+ | **PP-OCRv6** (detection + recognition) | Document OCR | PaddlePaddle (PaddleOCR) | Apache License 2.0 |
29
+
30
+ ### Attribution notes
31
+
32
+ - **Gemma models** are provided by Google under and subject to the **Gemma Terms
33
+ of Use** (ai.google.dev/gemma/terms) and the **Gemma Prohibited Use Policy**.
34
+ "Gemma is provided under and subject to the Gemma Terms of Use." Your use of the
35
+ Gemma models in emu is subject to those terms and that policy.
36
+ - **Qwen** models (Qwen 3.5, Qwen3-ASR, Qwen3-TTS) © Alibaba Cloud, licensed under
37
+ the Apache License 2.0.
38
+ - **gwen-tts** © G-Group AI Lab, MIT License; it is a fine-tune of Qwen3-TTS
39
+ (Apache-2.0).
40
+ - **Pocket TTS** © Kyutai, licensed under CC-BY-4.0 — attribution to Kyutai is
41
+ given here; weights are quantized/format-converted for on-device use.
42
+ - **VieNeu-TTS v3-Turbo** © pnnbao-ump, Apache-2.0; based on **MOSS-TTS-Nano** and
43
+ **MOSS-Audio-Tokenizer** © the OpenMOSS Team (Apache-2.0).
44
+ - **PP-OCRv6** © PaddlePaddle authors, Apache License 2.0.
45
+
46
+ Each model is redistributed/downloaded as quantized weights within emu's
47
+ self-contained bundle format and is used for on-device inference only.
48
+
49
+ ---
50
+
51
+ ## Software components
52
+
53
+ | Component | Use | License |
54
+ |---|---|---|
55
+ | **Qt 6** | Desktop UI framework (dynamically linked) | LGPL-3.0 (or commercial Qt license) |
56
+ | Qt 6 native binding layer (derived from **MIQT**, © mappu and contributors) | UI bindings | MIT License |
57
+ | **SQLite** | Local chat-history / settings storage | Public Domain |
58
+ | **FFmpeg** | Microphone / audio device capture | LGPL-2.1+ |
59
+ | **stb_image** | Image decoding | Public Domain / MIT |
60
+ | **NVIDIA CUDA Toolkit & cuDNN** (GPU builds only) | GPU acceleration runtime | NVIDIA Software License Agreement |
61
+
62
+ Qt 6 is used under the LGPL-3.0: it is dynamically linked, and the user may
63
+ replace the Qt runtime libraries. The NVIDIA CUDA/cuDNN runtimes are redistributed
64
+ in GPU builds under the NVIDIA Software License Agreement.
65
+
66
+ ---
67
+
68
+ ## Full license texts
69
+
70
+ The complete text of each referenced license — Apache License 2.0, the MIT
71
+ License, Creative Commons Attribution 4.0 International (CC-BY-4.0), the Gemma
72
+ Terms of Use and Prohibited Use Policy, LGPL-3.0 / LGPL-2.1, and the NVIDIA
73
+ Software License Agreement — accompanies the corresponding component and is
74
+ available from its provider. Copies are included with the distribution where
75
+ required by the applicable license.
76
+
77
+ © 2026 CloudKites Pty Ltd. This attribution file is provided to satisfy the
78
+ notice requirements of the licenses above and does not grant any rights in emu or
79
+ Numbat themselves, which remain proprietary (see `LICENSE`).