Model comparison
Qwen3-TTS-12Hz-1.7B-Base vs OmniVoice
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-Base | OmniVoice | |
|---|---|---|
| Parameters | 1.9B | 613M |
| Architecture | qwen3tts | omnivoice-lm |
| Layers | — | — |
| Native context | — | — |
| Mixture of experts | no | no |
| Quantizations published | 4 | 7 |
| Smallest quantization | 0.94 GiB | 0.56 GiB |
| Q4_K_M | — | 0.61 GiB |
| Licence | apache-2.0 | apache-2.0 |
KV cache by context
the term that decides long-context viability
| Context | Qwen3-TTS-12Hz-1.7B-Base | OmniVoice | Ratio |
|---|---|---|---|
| 4,096 | — | — | — |
| 8,192 | — | — | — |
| 16,384 | — | — | — |
| 32,768 | — | — | — |
| 65,536 | — | — | — |
| 131,072 | — | — | — |