Model comparison
Qwen3-TTS-12Hz-1.7B-VoiceDesign vs s2-pro
These two publish different quantization sets; the table below has the exact sizes.
From the file· summed bytes, KV per layer
Side by side
| Qwen3-TTS-12Hz-1.7B-VoiceDesign | s2-pro | |
|---|---|---|
| Parameters | 1.9B | 4.6B |
| Architecture | qwen3tts | fish-speech |
| Layers | — | 36 |
| Native context | — | — |
| Mixture of experts | no | yes, 1 experts |
| Quantizations published | 2 | 7 |
| Smallest quantization | 1.90 GiB | 2.40 GiB |
| Q4_K_M | — | 3.32 GiB |
| Licence | apache-2.0 | other |
KV cache by context
the term that decides long-context viability
| Context | Qwen3-TTS-12Hz-1.7B-VoiceDesign | s2-pro | Ratio |
|---|---|---|---|
| 4,096 | — | — | — |
| 8,192 | — | — | — |
| 16,384 | — | — | — |
| 32,768 | — | — | — |
| 65,536 | — | — | — |
| 131,072 | — | — | — |