Ming-3.0-tts-0.5B
Needs 1 GB at FP8 / INT8. The smallest real configuration today is 1× Quadro P2000 at Vast.ai, $0.03 per hour, about $22 a month at 730 hours.
- Per hour
- $0.03
- Per month
- $22
- Cheapest at
Vast.ai
What is documented about Ming-3.0-tts-0.5B
news checked 2026-10-27Dense Qwen2.5-0.5B backbone + DiT head; checkpoint holds 1.48B params including DiT head, audio tokenizer and speaker encoder. Chinese, English, German, Spanish, French, Korean, Japanese; voice design, dialects, free-form control. 'Ming-3.0-tts' ranked #3 (Elo 1189) on the Artificial Analysis open-weights TTS leaderboard read on 2026-10-27 (the listing does not say which size). Comfy-Org/Ming-3.0-TTS added 2026-10-23. Date = HF repo creation.
- https://huggingface.co/inclusionAI/Ming-3.0-tts-0.5B
- https://huggingface.co/api/models/inclusionAI/Ming-3.0-tts-0.5B?blobs=true
- https://artificialanalysis.ai/text-to-speech/leaderboard/provider-voice/open-weights
How much VRAM it needs
Weights plus 20% runtime overhead. For voice models, activations also depend on resolution, clip length and batch size; the per-GPU estimates below give the practical configuration.
| Precision | Weights | Weights + headroom |
|---|---|---|
| BF16published checkpoint | 3 GB | 4 GB |
| FP8 / INT8estimated | 1 GB | 1 GB |
| INT4 / MXFP4estimated | 0 GB | 1 GB |
What it costs to run, GPU by GPU
The fewest GPUs of one type whose total memory fits the model at short context, priced from the cheapest eligible on-demand offer for exactly that count (45-day window, newest reading per provider), with the next providers beside it. Monthly = 730 hours.
| GPU | Smallest config | VRAM used | $ / hour | $ / month | Cheapest at |
|---|---|---|---|---|---|
| Quadro P2000Best fit | 1× 5 GB5 GB total | 12% | $0.03 | $22 | |
| RTX 3060 | 1× 12 GB12 GB total | 5% | $0.04 | $26 | |
| RTX A2000 | 1× 12 GB12 GB total | 5% | $0.04 | $26 | |
| GTX 1660 Ti | 1× 6 GB6 GB total | 10% | $0.04 | $26 | |
| GTX 1050 | 1× 2 GB2 GB total | 30% | $0.04 | $27 | |
| Tesla P4 | 1× 8 GB8 GB total | 8% | $0.04 | $29 | |
| T4 | 1× 16 GB16 GB total | 4% | $0.04 | $29 | |
| GTX 1060 | 1× 6 GB6 GB total | 10% | $0.04 | $30 | |
| GTX 1660 | 1× 6 GB6 GB total | 10% | $0.04 | $30 | |
| Tesla P40 | 1× 24 GB24 GB total | 3% | $0.04 | $30 | |
| GTX Titan X | 1× 12 GB12 GB total | 5% | $0.04 | $31 | |
| Titan Xp | 1× 12 GB12 GB total | 5% | $0.04 | $31 | |
| GTX 1080 | 1× 8 GB8 GB total | 8% | $0.04 | $32 | |
| Tesla P100 | 1× 16 GB16 GB total | 4% | $0.04 | $33 | |
| RTX 2060 | 1× 6 GB6 GB total | 10% | $0.04 | $33 | |
| GTX 1650 | 1× 4 GB4 GB total | 15% | $0.05 | $35 | |
| GTX 1070 | 1× 8 GB8 GB total | 8% | $0.05 | $36 | |
| GTX 1660 Super | 1× 6 GB6 GB total | 10% | $0.05 | $38 | |
| Quadro P4000 | 1× 8 GB8 GB total | 8% | $0.06 | $40 | |
| RTX 2070 | 1× 8 GB8 GB total | 8% | $0.06 | $41 | |
| RTX 3060 Ti | 1× 8 GB8 GB total | 8% | $0.06 | $41 | |
| RTX 2060 Super | 1× 8 GB8 GB total | 8% | $0.06 | $41 | |
| GTX 1070 Ti | 1× 8 GB8 GB total | 8% | $0.06 | $44 | |
| RTX 2070 Super | 1× 8 GB8 GB total | 8% | $0.06 | $44 | |
| The 24 cheapest of 109 GPU types that fit. Every GPU, any precision and context, in the finder | |||||
