ZONOS2 (code, models, paper)

.gitattributes

@@ -33,3 +33,13 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
 *.zip filter=lfs diff=lfs merge=lfs -text
 *.zst filter=lfs diff=lfs merge=lfs -text
 *tfevents* filter=lfs diff=lfs merge=lfs -text
+models/ZONOS2-GGUF/dac.gguf filter=lfs diff=lfs merge=lfs -text
+models/ZONOS2-GGUF/spk-encoder.gguf filter=lfs diff=lfs merge=lfs -text
+models/ZONOS2-GGUF/zonos2-f16.gguf filter=lfs diff=lfs merge=lfs -text
+models/ZONOS2-GGUF/zonos2-q4_k.gguf filter=lfs diff=lfs merge=lfs -text
+models/ZONOS2-GGUF/zonos2-q5_k.gguf filter=lfs diff=lfs merge=lfs -text
+models/ZONOS2-GGUF/zonos2-q6_k.gguf filter=lfs diff=lfs merge=lfs -text
+models/ZONOS2-GGUF/zonos2-q8_0.gguf filter=lfs diff=lfs merge=lfs -text
+models/ZONOS2/assets/zonos2_arlooop_animated.gif filter=lfs diff=lfs merge=lfs -text
+models/ZONOS2/assets/ZONOS2BlogThumbnail.png filter=lfs diff=lfs merge=lfs -text
+ZONOS2[[:space:]]Technical[[:space:]]Report.pdf filter=lfs diff=lfs merge=lfs -text

ZONOS2 Technical Report.pdf

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3e25ed43bab8fa5c90d0d904f1ee56b93988dec759a067ed9823e80a4012ae2d
+size 2519958

code/ComfyUI-Zonos2-TTS.zip

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d00e3e275138ceeb2880a61ff1ef0d1ed74d5854dc7ab11378cd2a7b3dd15737
+size 99904

code/ZONOS2-mlx.zip

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b99ffd682c2a49fb0f2e05d9d80e6c7b9a9b7a9ecdcc6630dca42800ecba74e0
+size 3672748

code/ZONOS2.zip

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a944efd46dd3783ae78e6366fea602827ad5aed6a0badaec22a84799f8907b2e
+size 35462426

code/ZONOS2_finetuning.zip

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4d4572dd141821ebead3eb87aa579751cb8cd75f53f5675bf1acffecbcc570a8
+size 103003

code/Zonos2_TTS-ComfyUI.zip

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6710c41200587b6ba19463b097d36e35666ede0701b3524d875e3b6fdccdc8b7
+size 387522

code/mlx-zonos2.zip

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9d11a302ee72d551ebe75d255ec75281389003a29b0518c0cc52af4f9ec863d4
+size 420402

code/zonos2-dockerhub.zip

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:cad937a6702520fb9e6cfd96e9feca410f460cd441ba433aef5b4ccc2128e7fd
+size 32713

code/zonos2-webui.zip

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5f7bedb4d906c5d99cc0688602f11d63522ae2e0d63d3d48191ade5eb788e05b
+size 2215726

code/zonos2.cpp (MaurerAnton).zip

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f637e9028ef42faa7c48cef9df0fcc4fcfb99a3dd8c0ef0454ce0a20893187e8
+size 103309

code/zonos2.cpp.zip

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b07243268fb81ebf3494cc33d57a349aea15b1042ffe52c80a059831c87baf41
+size 338258436

models/ZONOS2-GGUF/.gitattributes

@@ -0,0 +1,45 @@
+*.7z filter=lfs diff=lfs merge=lfs -text
+*.arrow filter=lfs diff=lfs merge=lfs -text
+*.bin filter=lfs diff=lfs merge=lfs -text
+*.bz2 filter=lfs diff=lfs merge=lfs -text
+*.ckpt filter=lfs diff=lfs merge=lfs -text
+*.ftz filter=lfs diff=lfs merge=lfs -text
+*.gz filter=lfs diff=lfs merge=lfs -text
+*.h5 filter=lfs diff=lfs merge=lfs -text
+*.joblib filter=lfs diff=lfs merge=lfs -text
+*.lfs.* filter=lfs diff=lfs merge=lfs -text
+*.mlmodel filter=lfs diff=lfs merge=lfs -text
+*.model filter=lfs diff=lfs merge=lfs -text
+*.msgpack filter=lfs diff=lfs merge=lfs -text
+*.npy filter=lfs diff=lfs merge=lfs -text
+*.npz filter=lfs diff=lfs merge=lfs -text
+*.onnx filter=lfs diff=lfs merge=lfs -text
+*.ot filter=lfs diff=lfs merge=lfs -text
+*.parquet filter=lfs diff=lfs merge=lfs -text
+*.pb filter=lfs diff=lfs merge=lfs -text
+*.pickle filter=lfs diff=lfs merge=lfs -text
+*.pkl filter=lfs diff=lfs merge=lfs -text
+*.pt filter=lfs diff=lfs merge=lfs -text
+*.pth filter=lfs diff=lfs merge=lfs -text
+*.rar filter=lfs diff=lfs merge=lfs -text
+*.safetensors filter=lfs diff=lfs merge=lfs -text
+saved_model/**/* filter=lfs diff=lfs merge=lfs -text
+*.tar.* filter=lfs diff=lfs merge=lfs -text
+*.tar filter=lfs diff=lfs merge=lfs -text
+*.tflite filter=lfs diff=lfs merge=lfs -text
+*.tgz filter=lfs diff=lfs merge=lfs -text
+*.wasm filter=lfs diff=lfs merge=lfs -text
+*.xz filter=lfs diff=lfs merge=lfs -text
+*.zip filter=lfs diff=lfs merge=lfs -text
+*.zst filter=lfs diff=lfs merge=lfs -text
+*tfevents* filter=lfs diff=lfs merge=lfs -text
+dac.gguf filter=lfs diff=lfs merge=lfs -text
+spk-encoder.gguf filter=lfs diff=lfs merge=lfs -text
+zonos2-f16.gguf filter=lfs diff=lfs merge=lfs -text
+zonos2-q4_k-experts.gguf filter=lfs diff=lfs merge=lfs -text
+zonos2-q3_k-experts.gguf filter=lfs diff=lfs merge=lfs -text
+zonos2-q2_k-experts.gguf filter=lfs diff=lfs merge=lfs -text
+zonos2-q8_0.gguf filter=lfs diff=lfs merge=lfs -text
+zonos2-q6_k.gguf filter=lfs diff=lfs merge=lfs -text
+zonos2-q5_k.gguf filter=lfs diff=lfs merge=lfs -text
+zonos2-q4_k.gguf filter=lfs diff=lfs merge=lfs -text

models/ZONOS2-GGUF/README.md

@@ -0,0 +1,125 @@
+---
+license: apache-2.0
+pipeline_tag: text-to-speech
+library_name: gguf
+base_model: Zyphra/ZONOS2
+base_model_relation: quantized
+tags:
+  - text-to-speech
+  - tts
+  - gguf
+  - ggml
+  - zonos2.cpp
+---
+
+# ZONOS2 — GGUF
+
+<p align="center">
+  <img src="https://huggingface.co/Zyphra/ZONOS2/resolve/main/assets/ZONOS2BlogThumbnail.png" alt="ZONOS2 title card" width="750" />
+</p>
+
+<div align="center">
+  <a href="https://discord.gg/gTW9JwST8q" target="_blank">
+    <img src="https://img.shields.io/badge/Join%20Our%20Discord-7289DA?style=for-the-badge&logo=discord&logoColor=white" alt="Discord">
+  </a>
+</div>
+
+---
+
+GGUF weights for [**Zyphra/ZONOS2**](https://huggingface.co/Zyphra/ZONOS2), ready to run
+with [**zonos2.cpp**](https://github.com/Zyphra/zonos2.cpp) — a standalone ggml/GGUF C++
+port of ZONOS2. The **entire pipeline** (ECAPA speaker encoder → MoE backbone → DAC
+vocoder) runs as native C++ linking only `libggml` + `gguf`: **no Python, no PyTorch, no
+CUDA-only kernels at inference time.** CPU and CUDA use the same files.
+
+ZONOS2 is Zyphra's latest text-to-speech model, trained on more than 6 million hours of
+varied multilingual speech, delivering expressiveness and quality on par with—or even
+surpassing—top TTS providers at low latency with MoE. ZONOS2 excels at high-fidelity and
+naturalistic voice cloning.
+
+**For model details and speech samples, check out our [blog](https://www.zyphra.com/our-work/zonos2).**
+**A hosted version is available at [cloud.zyphra.com/audio-playground](https://cloud.zyphra.com/audio-playground).**
+
+Language support is as follows.
+| Tier   | Languages                                                                                                                                                                                      |
+| ------ | ---------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- |
+| Tier 1 | English, Mandarin Chinese, Japanese                                                                                                                                                            |
+| Tier 2 | Korean, Russian, Italian, Portuguese, French, Spanish, Vietnamese, German, Hebrew, Dutch                                                                                                       |
+| Tier 3 | Swedish, Hindi, Tamil, Telugu, Thai, Norwegian, Bengali, Tagalog, Arabic, Danish, Indonesian, Polish, Ukrainian, Romanian, Finnish, Hungarian, Lithuanian, Estonian, Slovak, Croatian, Latvian |
+
+---
+
+## Files
+
+| File | Size | Description |
+| --- | --- | --- |
+| `zonos2-f16.gguf` | 15.3 GB | F16 backbone — lossless from the bf16 checkpoint |
+| `zonos2-q8_0.gguf` | 8.5 GB | Q8_0 MoE experts, **F16 spine** — recommended; effectively lossless |
+| `zonos2-q6_k.gguf` | 6.8 GB | Q6_K MoE experts, F16 spine |
+| `zonos2-q5_k.gguf` | 5.9 GB | Q5_K MoE experts, F16 spine |
+| `zonos2-q4_k.gguf` | 4.9 GB | Q4_K MoE experts, F16 spine — smallest |
+| `dac.gguf` | 254 MB | DAC-44 kHz decoder (codes → waveform) |
+| `spk-encoder.gguf` | 24 MB | ECAPA-TDNN speaker encoder (wav → x-vector, for voice cloning) |
+
+All quants keep the **spine (attention + router + embeddings + speaker projection) at
+F16** and quantize only the MoE expert matrices — the layout that holds quality far below
+the usual quant floor. The full-precision spine is the single biggest quality lever; pick
+the expert precision (Q8 → Q4) that fits your VRAM.
+
+## Benchmarks
+
+| Build | Size | bpw | KLD&nbsp;↓ | Top-1&nbsp;↑ | WER&nbsp;↓ | SpkSim&nbsp;↑ | UTMOS&nbsp;↑ |
+| --- | --- | --- | --- | --- | --- | --- | --- |
+| **F16** (ref) | 15.3 GB | 16.0 | — | — | 2.79 | 66.75 | 4.40 |
+| **Q8_0** | 8.5 GB | 8.50 | 0.002 | 96.5% | 2.87 | 66.30 | 4.40 |
+| **Q6_K** | 6.8 GB | 6.56 | 0.007 | 92.9% | 3.07 | 66.12 | 4.40 |
+| **Q5_K** | 5.8 GB | 5.50 | 0.025 | 86.3% | 2.98 | 66.30 | 4.40 |
+| **Q4_K** | 4.9 GB | 4.50 | 0.072 | 76.9% | 3.00 | 64.54 | 4.36 |
+
+- **KLD** (mean) and **Top-1** measure how closely each quant tracks the F16 backbone's
+  per-frame logits, scored with the `zonos2-perplexity` tool in zonos2.cpp.
+- **WER** (Qwen3-ASR word-error rate), **SpkSim** (clone speaker similarity), and
+  **UTMOS** (predicted MOS) are end-to-end on the
+  [`Zyphra/ZTT1-Eval`](https://github.com/Zyphra/ZTT1-Eval) Clean English set;
+  lower WER and higher SpkSim/UTMOS are better.
+
+Although the logit metrics (KLD, Top-1) degrade steadily as the experts shrink, the
+**audio quality holds nearly flat down to Q4_K** — WER, speaker similarity, and UTMOS stay
+within eval noise of F16. The F16 spine keeps the model on-distribution, so the smaller
+quants spend their error budget on inaudible logit jitter rather than audible artifacts.
+**Q8_0 is the recommended default** (effectively lossless); Q4_K is a strong choice when
+VRAM is tight.
+
+## Quick Start
+
+Build [zonos2.cpp](https://github.com/Zyphra/zonos2.cpp) (CPU, or `-DGGML_CUDA=ON` for
+NVIDIA), then turn text into a waveform with one command:
+
+```bash
+zonos2-cli zonos2-q8_0.gguf --tts "Hello, world." out.wav \
+    --dac dac.gguf --gpu --seed 1
+```
+
+Add `--spk voice.mp3` (with `spk-encoder.gguf`) to clone a voice from a reference clip.
+
+### HTTP server
+
+`zonos2-server` mirrors the reference FastAPI — low-latency streaming PCM, an OpenAI
+`/v1/audio/speech` endpoint, in-process voice cloning, and a browser UI:
+
+```bash
+zonos2-server zonos2-q8_0.gguf --dac dac.gguf --spk-encoder spk-encoder.gguf --gpu
+```
+
+See the [zonos2.cpp README](https://github.com/Zyphra/zonos2.cpp) for build instructions,
+quantization (`quantize-cli`), batching, and the full CLI reference.
+
+## Citation
+If you find this model useful in an academic context please cite as:
+```
+@misc{zyphra2025zonos,
+  title     = {Zonos V2 Technical Report},
+  author    = {Gabriel Clark, Sofian Mejjoute, Mohamed Osman, George Close, Beren Millidge},
+  year      = {2026},
+}
+```

models/ZONOS2-GGUF/dac.gguf

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b5330fc112cfac5e7bae511a5083915ce5028899de4d7a4193dcc5874063532c
+size 254172448

models/ZONOS2-GGUF/languages.txt

@@ -0,0 +1,34 @@
+English
+Mandarin Chinese
+Japanese
+Korean
+Russian
+Italian
+Portuguese
+French
+Spanish
+Vietnamese
+German
+Hebrew
+Dutch
+Swedish
+Hindi
+Tamil
+Telugu
+Thai
+Norwegian
+Bengali
+Tagalog
+Arabic
+Danish
+Indonesian
+Polish
+Ukrainian
+Romanian
+Finnish
+Hungarian
+Lithuanian
+Estonian
+Slovak
+Croatian
+Latvian

models/ZONOS2-GGUF/source.txt

@@ -0,0 +1 @@
+https://huggingface.co/Zyphra/ZONOS2-GGUF

models/ZONOS2-GGUF/spk-encoder.gguf

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:284849980047a811e2c2a13ce7527c8a6a286021df152e5325533d783240eb6d
+size 24299552

models/ZONOS2-GGUF/zonos2-f16.gguf

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:01cdaa86fb74b7ecfec67267920d50235e4aaaa7b1faebb7254ae696ec72c3ff
+size 15336543104

models/ZONOS2-GGUF/zonos2-q4_k.gguf

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:dd78788653e3d3f668713cca9b27db4e564bae5f099609609be0a100e06a12f5
+size 4917891968

models/ZONOS2-GGUF/zonos2-q5_k.gguf

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7c1591c7232173700b70e05b3e44eda75f3f1596b455a4ea607a5ee748bf1ccc
+size 5823861632

models/ZONOS2-GGUF/zonos2-q6_k.gguf

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:349175207ce6521d4e89e2317f7862c68d15b8ee5ad876c1c3ce2938dc5c7e86
+size 6786454400

models/ZONOS2-GGUF/zonos2-q8_0.gguf

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:42dbdc642936bee4c9b0ed3045eadd815aba2166038d41b9abd04d683e911fb4
+size 8541770624

models/ZONOS2/.gitattributes

@@ -0,0 +1,38 @@
+*.7z filter=lfs diff=lfs merge=lfs -text
+*.arrow filter=lfs diff=lfs merge=lfs -text
+*.bin filter=lfs diff=lfs merge=lfs -text
+*.bz2 filter=lfs diff=lfs merge=lfs -text
+*.ckpt filter=lfs diff=lfs merge=lfs -text
+*.ftz filter=lfs diff=lfs merge=lfs -text
+*.gz filter=lfs diff=lfs merge=lfs -text
+*.h5 filter=lfs diff=lfs merge=lfs -text
+*.joblib filter=lfs diff=lfs merge=lfs -text
+*.lfs.* filter=lfs diff=lfs merge=lfs -text
+*.mlmodel filter=lfs diff=lfs merge=lfs -text
+*.model filter=lfs diff=lfs merge=lfs -text
+*.msgpack filter=lfs diff=lfs merge=lfs -text
+*.npy filter=lfs diff=lfs merge=lfs -text
+*.npz filter=lfs diff=lfs merge=lfs -text
+*.onnx filter=lfs diff=lfs merge=lfs -text
+*.ot filter=lfs diff=lfs merge=lfs -text
+*.parquet filter=lfs diff=lfs merge=lfs -text
+*.pb filter=lfs diff=lfs merge=lfs -text
+*.pickle filter=lfs diff=lfs merge=lfs -text
+*.pkl filter=lfs diff=lfs merge=lfs -text
+*.pt filter=lfs diff=lfs merge=lfs -text
+*.pth filter=lfs diff=lfs merge=lfs -text
+*.rar filter=lfs diff=lfs merge=lfs -text
+*.safetensors filter=lfs diff=lfs merge=lfs -text
+saved_model/**/* filter=lfs diff=lfs merge=lfs -text
+*.tar.* filter=lfs diff=lfs merge=lfs -text
+*.tar filter=lfs diff=lfs merge=lfs -text
+*.tflite filter=lfs diff=lfs merge=lfs -text
+*.tgz filter=lfs diff=lfs merge=lfs -text
+*.wasm filter=lfs diff=lfs merge=lfs -text
+*.xz filter=lfs diff=lfs merge=lfs -text
+*.zip filter=lfs diff=lfs merge=lfs -text
+*.zst filter=lfs diff=lfs merge=lfs -text
+*tfevents* filter=lfs diff=lfs merge=lfs -text
+zonos2_arlooop_animated.gif filter=lfs diff=lfs merge=lfs -text
+ZONOS2[[:space:]]Blog[[:space:]]Thumbnail.png filter=lfs diff=lfs merge=lfs -text
+assets/ZONOS2BlogThumbnail.png filter=lfs diff=lfs merge=lfs -text

models/ZONOS2/README.md

@@ -0,0 +1,121 @@
+---
+license: apache-2.0
+pipeline_tag: text-to-speech
+library_name: ZONOS2
+---
+
+# ZONOS2
+
+<p align="center">
+  <img src="./assets/ZONOS2BlogThumbnail.png" alt="ZONOS2 title card" width="750" />
+</p>
+
+<div align="center">
+  <a href="https://discord.gg/gTW9JwST8q" target="_blank">
+    <img src="https://img.shields.io/badge/Join%20Our%20Discord-7289DA?style=for-the-badge&logo=discord&logoColor=white" alt="Discord">
+  </a>
+</div>
+
+---
+
+
+ZONOS2 is our latest text-to-speech model trained on more than 6 million hours of varied multilingual speech, delivering expressiveness and quality on par with—or even surpassing—top TTS providers at low latency with MoE. ZONOS2 excels at high-fidelity and naturalistic voice cloning.
+
+
+During inference we use nemo TN normalized UTF-8 bytes and an ECAPA-TDNN embedding to generate DAC tokens with our MoE backbone. An inference overview can be seen below.
+<p align="center">
+  <img src="./assets/zonos2_arlooop_animated.gif" alt="ZONOS2 title card" width="750" />
+</p>
+
+Language support is as follows.
+| Tier   | Languages                                                                                                                                                                                      |
+| ------ | ---------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- |
+| Tier 1 | English, Mandarin Chinese, Japanese                                                                                                                                                            |
+| Tier 2 | Korean, Russian, Italian, Portuguese, French, Spanish, Vietnamese, German, Hebrew, Dutch                                                                                                       |
+| Tier 3 | Swedish, Hindi, Tamil, Telugu, Thai, Norwegian, Bengali, Tagalog, Arabic, Danish, Indonesian, Polish, Ukrainian, Romanian, Finnish, Hungarian, Lithuanian, Estonian, Slovak, Croatian, Latvian |
+
+
+For local inference we provide a high-performance TTS inference server built on [Mini-SGLang](https://github.com/sgl-project/mini-sglang).
+
+**For more details and speech samples, check out our [blog](https://www.zyphra.com/our-work/zonos2).**
+
+**We also have a hosted version available at [cloud.zyphra.com/audio-playground](https://cloud.zyphra.com/audio-playground).**
+
+---
+
+## Quick Start
+
+> **Platform Support**: Linux only (x86_64). Requires NVIDIA GPU with CUDA toolkit matching your driver version (`nvidia-smi` to check).
+
+### 1. Installation
+
+Requires [uv](https://docs.astral.sh/uv/getting-started/installation/).
+
+```bash
+git clone https://github.com/Zyphra/ZONOS2.git
+cd ZONOS2
+uv sync
+```
+
+### 2. Launch the TTS Server
+
+```bash
+uv run python -m minisgl --model-path Zyphra/ZONOS2 --tts-default-voices-dir ./default_voices/
+```
+
+`uv run` always uses the project environment, so no venv activation is needed.
+
+The server starts on `http://localhost:1919` by default. TTS mode is auto-detected for zonos2 models.
+`--tts-default-voices-dir <folder>` pre-populates the web UI with voice-clone
+speakers from disk; the folder is scanned recursively for speaker audio
+(`.wav`, `.mp3`, `.flac`, `.m4a`, `.ogg`, `.opus`, `.aac`, `.webm`) and saved
+embeddings (`.npy`, `.npz`). The newest voice is selected automatically on
+startup.
+
+### 3. Generate Speech
+
+**curl:**
+
+```bash
+curl -X POST http://localhost:1919/tts/generate \
+  -H "Content-Type: application/json" \
+  -d '{"text": "Hello world", "stream": true}' \
+  --output output.pcm
+
+# Convert to WAV
+ffmpeg -f f32le -ar 44100 -ac 1 -i output.pcm output.wav
+```
+
+**Web UI:** Open `http://localhost:1919/` in your browser.
+
+## Python API (offline inference)
+
+You can also run the engine directly in a Python script, without starting a
+server, via `TTSLLM`:
+
+```python
+from zonos2.message import TTSSamplingParams
+from zonos2.tts import TTSLLM
+
+tts = TTSLLM(model_path="Zyphra/ZONOS2")
+
+results = tts.generate(
+    ["Hello from the offline Python API.", "Batched prompts work too."],
+    TTSSamplingParams(seed=42),
+)
+
+for i, result in enumerate(results):
+    print(f"frames={len(result['audio_tokens'])}, eos_frame={result['eos_frame']}")
+    tts.save_audio(result["audio"], f"output_{i}.wav")
+```
+
+
+## Citation
+If you find this model useful in an academic context please cite as:
+```
+@misc{zyphra2025zonos,
+  title     = {Zonos V2 Technical Report},
+  author    = {Gabriel Clark, Sofian Mejjoute, Mohamed Osman, George Close, Beren Millidge},
+  year      = {2026},
+}
+```

models/ZONOS2/languages.txt

@@ -0,0 +1,34 @@
+English
+Mandarin Chinese
+Japanese
+Korean
+Russian
+Italian
+Portuguese
+French
+Spanish
+Vietnamese
+German
+Hebrew
+Dutch
+Swedish
+Hindi
+Tamil
+Telugu
+Thai
+Norwegian
+Bengali
+Tagalog
+Arabic
+Danish
+Indonesian
+Polish
+Ukrainian
+Romanian
+Finnish
+Hungarian
+Lithuanian
+Estonian
+Slovak
+Croatian
+Latvian

models/ZONOS2/model.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5f6aa0fff9036ee44ccbc625d40aa6bdd8ea223480a5447e9f6aad70c38b6ecd
+size 15336390655

models/ZONOS2/params.json

@@ -0,0 +1,59 @@
+{
+  "model_type": "zonos2",
+  "dtype": "bfloat16",
+  "n_layers": 28,
+  "dim": 2048,
+  "head_dim": 128,
+  "n_heads": null,
+  "n_kv_heads": 4,
+  "ffn_dim_multiplier": 1.5,
+  "multiple_of": 256,
+  "norm_eps": 1e-05,
+  "rope_theta": 10000.0,
+  "max_seqlen": 6144,
+  "n_codebooks": 9,
+  "codebook_size": 1024,
+  "eoa_id": 1024,
+  "audio_pad_id": 1025,
+  "text_vocab": 519,
+  "loss_softcap": 15.0,
+  "speaker_enabled": true,
+  "speaker_embedding_dim": 2048,
+  "speaker_lda_dim": 1024,
+  "speaker_background_token_enabled": true,
+  "accurate_mode_token_enabled": true,
+  "speaking_rate_num_buckets": 8,
+  "speaking_rate_buckets": ["0-8", "8-11", "11-14", "14-17", "17-21", "21-28", "28-40", "40+"],
+  "quality_num_buckets": 60,
+  "quality_features": [
+    "lufs",
+    "estimated_snr",
+    "max_pause",
+    "estimated_bandlimit_hz",
+    "leading_silence_s",
+    "trailing_silence_s"
+  ],
+  "quality_buckets": {
+    "lufs": ["-1000--50", "-50--45.5", "-45.5--41", "-41--36.5", "-36.5--32", "-32--27.5", "-27.5--23", "-23--18.5", "-18.5--14", "-14--9.5", "-9.5--5", "-5+"],
+    "estimated_snr": ["-1000-0", "0-6", "6-12", "12-18", "18-24", "24-30", "30-36", "36-42", "42-48", "48-54", "54-60", "60+"],
+    "max_pause": ["0-0.5", "0.5-1", "1-1.5", "1.5-2", "2-2.5", "2.5-3", "3-3.5", "3.5-4", "4-4.5", "4.5-5", "5-5.5", "5.5-6"],
+    "estimated_bandlimit_hz": ["495.3-3433", "3433-6371", "6371-9310", "9310-12248", "12248-15186", "15186-18124", "18124-21062", "21062-24000"],
+    "leading_silence_s": ["0-0.05", "0.05-0.1", "0.1-0.25", "0.25-0.5", "0.5-1", "1-2", "2-4", "4+"],
+    "trailing_silence_s": ["0-0.05", "0.05-0.1", "0.1-0.25", "0.25-0.5", "0.5-1", "1-2", "2-4", "4+"]
+  },
+  "quality_dropout": {
+    "lufs": 0.25,
+    "estimated_snr": 0.25,
+    "max_pause": 0.25,
+    "estimated_bandlimit_hz": 0.25,
+    "leading_silence_s": 0.25,
+    "trailing_silence_s": 0.25
+  },
+  "moe_impl": "sonic",
+  "moe_n_experts": 16,
+  "moe_router_topk": 1,
+  "special_topk_layers": {"26": 2},
+  "moe_router_dim": 128,
+  "moe_start_from_layer": 3,
+  "moe_end_from_layer": 1
+}

models/ZONOS2/source.txt

@@ -0,0 +1 @@
+https://huggingface.co/Zyphra/ZONOS2