drama-3
fishaudioOtheraudio

drama-3

Updated Nov 13, 2026
Parameters
35.9B
Context
262,144
License
Other
Updated
Nov 13, 2026

Will it run on your hardware?

Pick your GPU memory - see which quantizations fit, and the cheapest card for the rest

Runs on your 24 GB - best at Q4_K_M
1 of 4 quantizations fit drama-3 with real runtime overhead.
fits tight too big

Need an exact figure for your context length? Use the VRAM calculator.

Run it locally

Copy-paste - running in under a minute

vLLMOpenAI-compatible API
vllm serve fishaudio/drama-3

New to this? Start with Ollama · serve to many users with vLLM.

Deep dive

Notes, sources, and the full write-up

drama-3 is an open-weight text-to-speech model from fishaudio, published on Hugging Face in October 2026. It has 35.9B parameters and is released under a custom license (see the model card).

HardwareLink to this section

At full FP16 precision the weights need about 72 GB of memory; 8-bit quantization brings that to roughly 38 GB. Check the VRAM tiers above to see which GPUs fit it.

PopularityLink to this section

drama-3 had 48,213 downloads on Hugging Face in the last month and has 1,262 likes.

Run it locallyLink to this section

The weights are on Hugging Face. Follow the model card for the exact setup and license terms.

Frequently asked

Quick answers to common questions

How much VRAM does drama-3 need?

drama-3 with 35.9B parameters needs approximately 21 GB at Q4_K_M quantization. Use our VRAM calculator for an exact estimate.

Is drama-3 better than other fishaudio models?

drama-3 has 35.9B parameters with 262,144 context - a strong choice for text-to-speech.

What license is drama-3 under?

drama-3 is released under the Other license, making it suitable for most commercial and personal projects.

What hardware runs drama-3 well?

With 35.9B parameters, drama-3 requires adequate VRAM. High-end GPUs like the RTX 4090 (24GB), RTX 5090 (32GB), or Mac Studio with unified memory are good options. Check our hardware directory for specific recommendations.

What is the best quantization for drama-3?

Q4_K_M is the recommended sweet spot - ~98% of FP16 quality at ~27% of the size. Q5_K_M (~25 GB) is an option if you have spare VRAM. Use our VRAM calculator to compare.

How long can drama-3's context window handle?

drama-3 supports a 262,144-token context window - enough for very long documents, codebases, or multi-turn conversations. Real-world usable context may vary by implementation.

What models compete with drama-3?

drama-3 competes with other 18B–54B. Browse our model directory for comparisons, benchmarks, and community reviews to find the best fit.

Nearby options

Similar models and compatible hardware by spec