ZONOS2 GGUF

Quantization: Q8_0 is the floor for this architecture — smaller quants (Q6_K and below) are incoherent, because ZONOS2's SonicMoE residual structure amplifies low-bit error. Use Q8_0 or F16.

Usage (mistral.rs)

Serve the model:

mistralrs-server -p 8080 speech --arch zonos2 --model-id ZONOS2-Q8_0.gguf

Generate speech via the OpenAI-compatible /v1/audio/speech endpoint. For voice cloning, pass a speaker embedding:

curl localhost:8080/v1/audio/speech \
  -H 'Content-Type: application/json' \
  -d '{
        "model": "default",
        "input": "Your text here.",
        "speaker_embedding": [ /* 2048-dim speaker vector */ ],
        "response_format": "wav"
      }' --output out.wav

• speaker_embedding — the [2048]-dim speaker vector (e.g. from an ECAPA speaker encoder). Omit it for unconditional generation.
• Tip: punctuate naturally — short clauses help prosody and clean utterance termination.

Model

• Architecture: ZONOS2 (Zyphra) — SonicMoE decoder + DAC neural codec.
• Output: 16-bit mono PCM, 44.1 kHz.
• Voice cloning: via a per-request speaker embedding.

Provenance

Converted from the upstream Zyphra/ZONOS2 checkpoint with mistralrs-modeltool zonos2-gguf.

--- license: other base_model: Zyphra/ZONOS2 pipeline_tag: text-to-speech tags: - gguf - tts - text-to-speech - voice-cloning - mistral.rs - sonicmoe --- # ZONOS2-GGUF GGUF builds of [Zyphra/ZONOS2](https://huggingface.co/Zyphra/ZONOS2) — an ~8B-parameter (SonicMoE, ~900M active) real-time text-to-speech model with voice cloning — for [mistral.rs](https://github.com/EricLBuehler/mistral.rs). ## Files | File | Size | Use | |------|------|-----| | `ZONOS2-Q8_0.gguf` | 8.27 GB | **Recommended.** Near-lossless and the smallest coherent quantization. | | `ZONOS2-F16.gguf` | 15.3 GB | Full precision, for maximum fidelity / reference. | > **Quantization:** Q8_0 is the floor for this architecture — smaller quants (Q6_K and below) are incoherent, because ZONOS2's SonicMoE residual structure amplifies low-bit error. Use **Q8_0** or **F16**. ## Usage (mistral.rs) Serve the model: ```bash mistralrs-server -p 8080 speech --arch zonos2 --model-id ZONOS2-Q8_0.gguf ``` Generate speech via the OpenAI-compatible `/v1/audio/speech` endpoint. For voice cloning, pass a speaker embedding: ```bash curl localhost:8080/v1/audio/speech \ -H 'Content-Type: application/json' \ -d '{ "model": "default", "input": "Your text here.", "speaker_embedding": [ /* 2048-dim speaker vector */ ], "response_format": "wav" }' --output out.wav ``` - `speaker_embedding` — the `[2048]`-dim speaker vector (e.g. from an ECAPA speaker encoder). Omit it for unconditional generation. - Tip: punctuate naturally — short clauses help prosody and clean utterance termination. ## Model - **Architecture:** ZONOS2 (Zyphra) — SonicMoE decoder + DAC neural codec. - **Output:** 16-bit mono PCM, 44.1 kHz. - **Voice cloning:** via a per-request speaker embedding. ## Provenance Converted from the upstream Zyphra/ZONOS2 checkpoint with `mistralrs-modeltool zonos2-gguf`.