Drama-3-GGUF
--- basemodel: fishaudio/drama-3 basemodelrelation: quantized pipelinetag: text-to-speech license: other licensename: fish-audio-community-license-v1.0 licenselink: https://huggingface.co/fishaudio/drama-3/blob/main/LICENSE tags: - gguf - audio.cpp - fish-audio - text-to-speech - voice-cloning - moe - mixed-quantization ---
Params
—
Context
—
Downloads 30d
496 K
Likes
39
Download history
daily snapshots · 9 days496 K383 K
Nov 2Nov 5Nov 8Nov 10
Can you run it?
Estimated VRAM at 8K context unless noted. Pick your hardware to see the verdict per quantization.
| File | Quant | Size | Est. VRAM | Verdict on RTX 4090 · 24 GB |
|---|---|---|---|---|
| drama-3-backbone-8bit-experts-Q2_K.gguf | Q2_K | 17.1 GB | 19.3 GB | ✅ Runs comfortably |
| drama-3-backbone-8bit-experts-Q3_K_S.gguf | Q3_K_S | 20.3 GB | 22.9 GB | ⚠️ Tight — reduce context |
| drama-3-backbone-8bit-F32-routers-BF16-embed-head-experts-Q4_K_S.gguf | Q4_K_S | 25.6 GB | 28.7 GB | ❌ Won’t fit |
Estimate: file size × 1.1 + KV cache at 8K + 0.5 GB overhead. Not a benchmark — how we calculate this.
Specifications
- Licence
- other
- First seen on the Hub
- 2026-10-13
- Training datasets
- undisclosed
- Added to our catalog
- 2026-11-02
Compare with any text-to-speech model
Popular comparisons
Appears in