SAVRN Model Hub · Comparisons
Qwen3-TTS-12Hz-1.7B-CustomVoice vs VoxCPM2
Qwen3-TTS-12Hz-1.7B-CustomVoice has 1.9B parameters and VoxCPM2 has 2.3B parameters; both are released under Apache License 2.0; at 16-bit, Qwen3-TTS-12Hz-1.7B-CustomVoice needs about 4.6 GB (1x MI300X from $1.85 an hour) and VoxCPM2 about 5.5 GB (1x MI300X from $1.85 an hour).
| Field | Qwen3-TTS-12Hz-1.7B-CustomVoice Qwen/Qwen3-TTS-12Hz-1.7B-CustomVoice | VoxCPM2 openbmb/VoxCPM2 |
|---|---|---|
| Publisher | Qwen | OpenBMB |
| Task | Text to speech | Text to speech |
| Modality | Audio | Audio |
| Parameters, as reported | 1.9B parameters | 2.3B parameters |
| Architecture | Qwen3TTSForConditionalGeneration | Not stated |
| Library | Not stated | voxcpm |
| Context length | Not stated | Not stated |
| Repository size | 4.5 GB | 5.0 GB |
| Artifact formats | safetensors | safetensors |
| License | apache-2.0 | apache-2.0 |
| Access | Open weights, no gate | Open weights, no gate |
| Memory at 16-bit (weights and margin) | 4.6 GB | 5.5 GB |
| Cheapest GPUs at 16-bit, per hour | 1x MI300X, $1.85 | 1x MI300X, $1.85 |
| Memory at 4-bit (weights and margin) | 1.2 GB | 1.4 GB |
| Cheapest GPUs at 4-bit, per hour | 1x MI300X, $1.85 | 1x MI300X, $1.85 |
| Revision viewed | 0c0e3051f131 | 32279effe8c1 |
| Downloads reported by the hub | 2.6M | 385.5k |
| Last observed | 2026-09-18 | 2026-09-18 |
An evaluation row appears only where at least two of these models report the same benchmark with the same stated configuration, metric, unit and setup. Different evaluators stay named in each cell. Values are shown as reported: no unit conversion, no ranking.
SAVRN's Notes on VoxCPM2
Where does a 2.3 billion parameter text-to-speech model belong in a facility built around 192 GB accelerators? Beside something bigger. VoxCPM2 needs 5.5 GB at 16-bit, 2.7 GB at 8-bit and 1.4 GB at 4-bit, and the cheapest priced slot, one MI300X at $1.85 an hour, has room for many copies. It outputs 48 kHz audio in 30 languages with no language tag, designs a voice from a written description, and clones one from a short clip.
Apache 2.0 keeps the deployment conversation short: commercial use, modification and redistribution are permitted, provided the license, copyright notices and any NOTICE file stay with the files and significant changes are stated. Two checks before committing: the runtime is its own voxcpm library rather than transformers, so plan the serving path around it, and no context length is recorded. The paper is arXiv:2509.24650; the files were last updated 2026-08-18.
Questions
Which is larger, Qwen3-TTS-12Hz-1.7B-CustomVoice or VoxCPM2?
VoxCPM2 (2.3B parameters) is larger than Qwen3-TTS-12Hz-1.7B-CustomVoice (1.9B parameters), by the parameter counts their publishers report.
Which is cheaper to run, Qwen3-TTS-12Hz-1.7B-CustomVoice or VoxCPM2?
At 4-bit, Qwen3-TTS-12Hz-1.7B-CustomVoice fits on 1x MI300X from $1.85 an hour and VoxCPM2 on 1x MI300X from $1.85 an hour, at the lowest on-demand prices the SAVRN Index lists.
Can I use Qwen3-TTS-12Hz-1.7B-CustomVoice commercially?
Yes. Qwen3-TTS-12Hz-1.7B-CustomVoice is released under Apache License 2.0. The Apache License 2.0 is a permissive open-source license. It permits commercial use, modification and redistribution. It requires keeping the license and copyright notices and any NOTICE file, stating significant changes, and it includes an express patent grant from contributors.
Can I use VoxCPM2 commercially?
Yes. VoxCPM2 is released under Apache License 2.0. The Apache License 2.0 is a permissive open-source license. It permits commercial use, modification and redistribution. It requires keeping the license and copyright notices and any NOTICE file, stating significant changes, and it includes an express patent grant from contributors.