cstr/breeze-tts-2-GGUF

Breeze TTS 2 — GGUF (CrispASR)

Derived from Breeze TTS 2 by BreezeBlue and licensed for research and non-commercial use only.

⚠️ NON-COMMERCIAL. These files are Derivative Models under §1.3 of the BreezeBlue Research and Non-Commercial License Agreement, which names quantization explicitly. They inherit the licence of the original weights. Hosting them as a service is a commercial use under §1.7(b). The full Agreement is in LICENSE; the required notice is in NOTICE.

GGUF conversion of BreezeBlue/Breeze-TTS-2 for CrispASR.

File	Size	Notes
breeze-tts-2-f16.gguf	~5.7 GB	reference precision
breeze-tts-2-q8_0.gguf	~3.0 GB
breeze-tts-2-q4_k.gguf	~2.2 GB	default for the CrispASR registry

Architecture: a CSM fork — T5Gemma2 text encoder (26L, bidirectional, symmetric sliding window) → Qwen3 backbone (28L) → depth decoder (12L) over 16 codebooks at 12.5 Hz. Languages: English and Chinese.

The codec is not in these files. Breeze's bundled audio tokenizer is bit-identical to Qwen/Qwen3-TTS-Tokenizer-12Hz, which CrispASR already ships as cstr/qwen3-tts-tokenizer-12hz-GGUF. It is wired in as a registry companion and downloaded alongside the model. Roughly 1.16 GB of the original checkpoint (embed_text_tokens and a leftover Mimi codec_model.*) is unreachable at inference and is dropped by the converter.

Inference code in the upstream repo is Apache-2.0 and is not covered by the Agreement (§1.2); the restriction here is on the weights.

All credit for the model goes to BreezeBlue / RESONIA, INC.