---
title: Breeze-TTS-2-comfyui
canonical_url: "https://www.modelscope.cn/models/drbaph/Breeze-TTS-2-comfyui"
md_url: "https://www.modelscope.cn/models/drbaph/Breeze-TTS-2-comfyui.md"
repository: drbaph/Breeze-TTS-2-comfyui
last_updated: 2026-09-14
license: other
pipeline_tag: text-to-speech
tasks:
  - text-to-speech
model_type:
  - breeze
architectures:
  - BreezeForConditionalGeneration
base_model:
  - BreezeBlue/Breeze-TTS-2
base_model_relation: finetune
parameters: 14.1B
tensor_type:
  - BF16
  - F32
  - I8
  - U8
library_name:
  - safetensors
  - pytorch
frameworks:
  - pytorch
language:
  - en
  - zh
downloads: 815
stars: 3
tags:
  - text-to-speech
  - transformers
  - safetensors
  - pytorch
  - cuda
  - english
  - chinese
  - breeze
  - text-generation
  - speech-generation
  - voice-clone
  - voice-design
  - voice-direction
  - comfyui
  - int8
  - convrot
  - quantization
  - quantised
  - hadamard
---

# Breeze-TTS-2-comfyui

> Breeze-TTS-2-comfyui - drbaph 在 ModelScope 开源的模型。Breeze TTS 2 — ComfyUI derivatives

drbaph/Breeze-TTS-2-comfyui 是 ModelScope 魔搭社区上的 14.1B 参数text-to-speech模型，采用 other 许可，基于 BreezeBlue/Breeze-TTS-2 构建。

- **Repository**: drbaph/Breeze-TTS-2-comfyui
- **License**: other
- **Tasks**: text-to-speech
- **Parameters**: 14.1B
- **Base model**: BreezeBlue/Breeze-TTS-2
- **Tags**: text-to-speech, transformers, safetensors, pytorch, cuda, english, chinese, breeze, text-generation, speech-generation, voice-clone, voice-design, voice-direction, comfyui, int8, convrot, quantization, quantised, hadamard
- **Downloads**: 815
- **Stars**: 3
- **Last updated**: 2026-09-14

Source: https://www.modelscope.cn/models/drbaph/Breeze-TTS-2-comfyui

---

# Breeze TTS 2 — ComfyUI derivatives

Converted from [BreezeBlue/Breeze-TTS-2](https://huggingface.co/BreezeBlue/Breeze-TTS-2) for the
[ComfyUI-Breeze-TTS-2](https://github.com/saganaki22/ComfyUI-Breeze-TTS-2) node pack.


![Screenshot 2026-08-26 160830](https://cdn-uploads.huggingface.co/production/uploads/63473b59e5c0717e6737b872/GIenR6hWJl1eABum71BEy.png)

## Contents

| File | Description |
| --- | --- |
| `Breeze-TTS-2-bf16.safetensors` | The official checkpoint merged from 2 shards into a single file (bit-exact, bf16, 6.49 GiB). |
| `Breeze-TTS-2-int8-convrot.safetensors` | All 462 transformer projection linears INT8 ConvRot (backbone + text encoder + depth decoder), comfy-kitchen per-channel per-row scales. Smallest build. |
| `Breeze-TTS-2-int8-hybrid.safetensors` | Backbone + text encoder INT8 ConvRot; the depth decoder (the per-frame autoregressive hot loop) stays bf16. **Recommended INT8 build: bf16-level speed with −27 % VRAM.** |
| `Breeze-TTS-2-int8-text-encoder.safetensors` | Text encoder only INT8; everything decode-time stays bf16. |
| `audio_tokenizer/` | Unmodified Qwen3-TTS 12 Hz audio tokenizer (codec) used for reference-audio encoding and speech decoding. |
| `quantization_report_*.json` | Per-tensor dequantization error statistics for each INT8 build. |
| `benchmark_report.json` | RTF / VRAM / log-mel / Whisper-WER comparison across all four builds. |

## Benchmarks (RTX 5090, ComfyUI loader, sdpa, 24 kHz output)

| Build | avg RTF (lower = faster) | Peak VRAM | Whisper WER vs prompt* |
| --- | --- | --- | --- |
| bf16 | 5.68 | 7.51 GiB | baseline |
| int8-convrot | 9.08 (+60 %) | 5.21 GiB (−31 %) | matches bf16 |
| **int8-hybrid** | **5.64 (parity)** | **5.53 GiB (−27 %)** | matches bf16 |
| int8-text-encoder | 5.34 | 6.82 GiB (−9 %) | matches bf16 |

*All builds including bf16 share the same few ASR-level blurs on the clone prompt; design/direction
prompts transcribe identically to bf16 on the hybrid builds.

The depth decoder runs 15 skinny single-token GEMMs per 12.5 Hz frame (≈1260 linear calls/frame),
where int8 per-call overhead (quantize + dispatch, ~0.1 ms/call on Windows) outweighs the GEMM
savings — that is why full INT8 regresses and the hybrid does not.

## License

- **Weights**: the Breeze TTS 2 model weights, these derivative files, and self-hosted outputs are
  governed by the BreezeBlue Research and Non-Commercial License (see `license_link`). Commercial
  use requires written authorization from RESONIA, INC.
- **Node-pack code**: [ComfyUI-Breeze-TTS-2](https://github.com/saganaki22/ComfyUI-Breeze-TTS-2)
  is Apache-2.0. The Breeze TTS source code is also Apache-2.0.