Qwen3-TTS-12Hz-1.7B-CustomVoice vs VibeVoice-Realtime-0.5B
Specs, VRAM requirements and download trends — updated 18 September 2026.
Qwen · A · Open weights
Params
1.9 B
Context
—
30d
1.4 M
apache-2.0 · commercial OK
microsoft · B · Open weights
Params
1.0 B
Context
—
30d
385 K
mit · commercial OK
Specification comparison
Differences are highlighted; identical values are muted.
| Specification | Qwen3-TTS-12Hz-1.7B-CustomVoice | VibeVoice-Realtime-0.5B |
|---|---|---|
| Parameters | 1.9 B | 1.0 B |
| Architecture | Qwen3TTSForConditionalGeneration | VibeVoiceStreamingForConditionalGenerationInference |
| Context length | — | — |
| Licence | apache-2.0 | mit |
| Commercial use | Allowed | Allowed |
| Languages | — | 1 |
| Downloads 30d | 1,412,025 | 384,773 |
| Downloads all time | 18.2 M | 5.8 M |
| Quantizations on the Hub | SAFETENSORS | SAFETENSORS |
| Gated | No | No |
| First seen on the Hub | 2026-01-21 | 2025-12-04 |
Download trend
Daily snapshots, last 90 days (13 Aug – 10 Nov)
2.6 M385 K
Qwen3-TTS-12Hz-1.7B-CustomVoice
VibeVoice-Realtime-0.5B
VRAM side by side
On RTX 4090 · 24 GB · 8K context unless noted
VibeVoice-Realtime-0.5B · fp16
2.9 GB / 24 GB
Qwen3-TTS-12Hz-1.7B-CustomVoice · fp16
5.8 GB / 24 GB