Speech-to-text
Official
distil-large-v3
Listed in tt-inference-server on 6 hardware types; best status Complete.
In tt-inference-server
| Hardware | Engine and implementation | Status | Context | Users | Build | Support page |
|---|---|---|---|---|---|---|
BH QuietBox 2P300X2 |
media whisper |
Complete | 64K | 4 | tt-metal 25891d3Image ghcr.io/tenstorrent/tt-media-inference-server:0.15.0-25891d3 |
- |
WH GalaxyGALAXY |
media whisper |
Complete | 64K | 32 | tt-metal 8c48a10Image ghcr.io/tenstorrent/tt-media-inference-server:0.17.0-8c48a10 |
- |
WH LoudBox/QuietBoxT3K |
media whisper |
Complete | 64K | 4 | tt-metal bac8b34Image ghcr.io/tenstorrent/tt-media-inference-server:0.11.1-bac8b34 |
- |
n150N150 |
media whisper |
Complete | 64K | 1 | tt-metal bac8b34Image ghcr.io/tenstorrent/tt-media-inference-server:0.11.1-bac8b34 |
- |
n300N300 |
media whisper |
Complete | 64K | 1 | tt-metal bac8b34Image ghcr.io/tenstorrent/tt-media-inference-server:0.11.1-bac8b34 |
- |
p150P150 |
media whisper |
Complete | 64K | 1 | tt-metal 25891d3Image ghcr.io/tenstorrent/tt-media-inference-server:0.15.0-25891d3 |
- |
Status, context and users are copied from tt-inference-server's release spec (commit 2db98e38c, release 0.25.0). Context is the longest prompt plus reply the entry allows. Users is its highest concurrency. Each row carries its own code link, tt-metal and vLLM commits and image, because they differ between hardware. Nothing on this page is measured.
tt-model bundles
No tt-model bundle has been published for this model.
Details
| Kind | Speech-to-text (from tt-inference-server model type) |
| Family | distil (from the start of the name) |
| Implementation | whisper |
| Disk needed | 15 GB |
| RAM needed | 6 GB |
| Listed by | tt-inference-server |