Model reference · open weights
stable-audio-3-medium is an open-weight audio or speech model from thepatch, listed in the AxForge catalogue. AxForge can bring it up on EU-owned hardware for you on request — with the licence handled where one is required.
About
Stable Audio 3 Medium — GGUF (for sa3.cpp) GGUF conversions of stabilityai/stable-audio-3-medium for sa3.cpp — a portable C++/GGML port of Stable Audio 3, no PyTorch in the loop. Runs on CPU, CUDA, Vulkan, or Metal (Apple Silicon). Every component is validated against the PyTorch reference at cosine similarity ~1.0. Files This is a multi-file model. Grab the DiT + SAME at your chosen precision and the conditioner, plus the shared encoder + tokenizer from the t5gemma-b-b-ul2-GGUF repo. F16 is the production path (~3.5s for 12s of audio on an 8GB laptop GPU); F32 is for CPU validation. The conditioner + encoder + tokenizer stay F32 (small / quality-critical). Encodings sa3-generate --encoding resolves the DiT and the SAME with the same suffix, so download the pair. q4km and q5km promote the attention V, feed-forward down and embedding tensors to Q6K; q80 is uniform. Every tier passes sa3-quant-check with below-threshold=0 at cosine 0.990 against the F16 reference, for the DiT and the SAME alike. Quantization buys footprint everywhere and speed only on some backends. CUDA and Vulkan gain roughly 33% end to end. Metal is flat — Q80 is 1.7% faster and Q4KM 2.1% slower than F16, because the load-time saving and the added per-step dequant cancel out. On a Mac, pick a quant for the memory, not for the speed. Usage For use with sa3.cpp: For a quantized set, pass the encoding to both — the downloader and the generator use the same names: Performance Roughly 3s for a 12s clip at f16 on an 8GB laptop GPU (RTX 5070), and ~6s on an Apple M4 — end to end, including model load. The sliding-window decoder keeps long generations linear (a 2-minute clip is ~9s on the 5070). CPU works but is ~10× slower. Full numbers + the f16 / flash-attention levers: docs/BENCHMARKS.md. License These are format conversions of stabilityai/stable-audio-3-medium, whose weights Stability AI releases under the Stability AI Community License: free for organizations under $1M annual revenue, with commercial use, fine-tuning, and derivative works permitted within that threshold (above it, contact Stability AI for an Enterprise License). Outputs are yours. That license carries over to these converted weig
Summarised from the published model card. Read the full card on the HuggingFace links below.
Specifications
| Maker | thepatch |
|---|---|
| Type | Audio & music |
| Variants | 1 |
| Based on | stabilityai/stable-audio-3-medium |
| Released | 2026-06-29 |
| Popularity | 2k downloads / month |
| Likes | 4 |
| Licence | Commercial licence needed |
How it works
Variants
Open weights ship in several sizes and precisions. One page, all the variants — pick the one that fits your GPU. VRAM figures are estimates from model size.
| Variant | Params | Precision | VRAM | Fits 16 GB | Weights |
|---|---|---|---|---|---|
| stable-audio-3-medium-GGUF | — | GGUF | — | — | Weights ↗ |
Using it via the API
Once AxForge deploys thepatch-stable-audio-3-medium for you, it answers on the OpenAI-compatible API — the same base URL and keys as every other model. (thepatch-stable-audio-3-medium below is illustrative; you get the exact model name on deployment.)
$ curl -sS https://api.axforge.ai/v1/audio/transcriptions \ -H "Authorization: Bearer $AXFORGE_API_KEY" \ -F model="thepatch-stable-audio-3-medium" -F file=@audio.mp3
Details
Languages
Tags
Licence
The weights are open but its licence needs a commercial agreement for business use. AxForge can arrange that licence and host the model for you — you pay AxForge, we settle with the model’s maker. Ask us for a quote. Read the licence ↗