> This page is for NVIDIA Studio Voice NIM, version 1.2.0.
> For other versions, use one of these documentation indexes:
> - 1.2.1 (Latest) (default): https://docs.nvidia.com/nim/maxine/studio-voice/1.2.1/llms.txt
> - 1.2.0: https://docs.nvidia.com/nim/maxine/studio-voice/1.2.0/llms.txt
> - 1.1.0: https://docs.nvidia.com/nim/maxine/studio-voice/1.1.0/llms.txt
> - 1.0.0: https://docs.nvidia.com/nim/maxine/studio-voice/1.0.0/llms.txt

> For clean Markdown of any page, append .md to the page URL.
> For a complete documentation index, see https://docs.nvidia.com/nim/maxine/llms.txt.

# NIM Model Profile Table

Each `NIM_MODEL_PROFILE` downloads a specific Model Type on the supported GPU.

<Note>
If `NIM_MODEL_PROFILE` is not set, the NIM automatically selects a matching `NIM_MODEL_PROFILE` based on the target GPU architecture; that is, SM (Streaming Multiprocessor) values.
</Note>

## Model Type

| Model Type | Description |
|------------|-------------|
| 48k-HQ     | High Quality Model that supports input audio of sample rate 48KHz |
| 16k-HQ     | High Quality Model that supports input audio of sample rate 16KHz |
| 48k-LL     | Streaming-compatible Model that supports input audio of sample rate 48KHz    |

## NIM Model Profile Table

| GPU                  |  SM  | Model Type | NIM_MODEL_PROFILE                                                |
|----------------------|------|------------|------------------------------------------------------------------|
| T4                   |  7.5 | 16k-HQ     | 669026efdaf09bc9fb3bcfc63761ae981f588bafc2e17c201f3747484f405ff2 |
| T4                   |  7.5 | 48k-HQ     | a81adf7c20f5c53ad6622afd38768a980bc62f536dba16cdff32e03505fd3044 |
| T4                   |  7.5 | 48k-LL     | d4a05fd85322c89b2593de7bb90dff8f51c6e065476c127abdd3de7bc94a29cc |
| A30, A100            |  8.0 | 16k-HQ     | 333b238d3b4e5393b7c3ef5da4321027a6c89a24f22cda08eefb6b02aae31a12 |
| A30, A100            |  8.0 | 48k-HQ     | a0ef5f245313252c7c48f5ba2660b7457e388a6ac6b51c24c87845a1ec3498b3 |
| A30, A100            |  8.0 | 48k-LL     | 49ed404d6a0ce29e83ed33cc3cfb9a4c0dbb3ca200c53743efc9f57a45c95b92 |
| A2, A10, A16, A40    |  8.6 | 16k-HQ     | 26fa7ba0f7f609439db66983da76e5863958a5367096e111ec33ada2a3f62b4d |
| A2, A10, A16, A40    |  8.6 | 48k-HQ     | 0cd76f5d9cbfc7054ed4c7164dcfb9fa2427963f4df80ec3d5ccb80db2311b9b |
| A2, A10, A16, A40    |  8.6 | 48k-LL     | 0cc07261902d91a87f12dff8c5b028b49b98afafc95130d61b069141ac021be6 |
| L4, L40, RTX 4090    |  8.9 | 16k-HQ     | fc6090cd82ff3e918cba9e6b82b1bb72932ce4ec45979584f2a2bd248429302d |
| L4, L40, RTX 4090    |  8.9 | 48k-HQ     | b4f601e14bb68d8f11f2f6ad38ed84e150fb197653f3125362aab9a9e1c7fc20 |
| L4, L40, RTX 4090    |  8.9 | 48k-LL     | 32ad9bfbc222aa6f74c62898be9269c81523a7bc877fab7cb21268f47ef1fbfb |
| H100                 |  9.0 | 16k-HQ     | acc77362f0a2b0570a618caf97bcb9d45d8d64e18b2c79098a940ea88afc932c |
| H100                 |  9.0 | 48k-HQ     | a0fb1bacd8fb1b15a22aee2809cae8c8cac83ddf43681ab7f49720363d7f9606 |
| H100                 |  9.0 | 48k-LL     | da9c18295abb1cf2b273a6fc48ed9633d45872fc1ba5a344d9aaf15c1c83a893 |
| B100, B200           | 10.0 | 16k-HQ     | 1ce3a6a13c0eeffab093af91d26c36bdf1dd01c6bab85001514f9fc930c8c6a6 |
| B100, B200           | 10.0 | 48k-HQ     | 937b741a8b77ae95ba009b63839991c1f3ab4007bab67398adde9ace57d0edaf |
| B100, B200           | 10.0 | 48k-LL     | e9ae7b9757846cb5f76d23b345d5d2a4691876d32e13ce661e51735e4a4583e1 |
| B40, RTX 5090, RTX 5080, RTX PRO 6000 Blackwell Server Edition | 12.0 | 16k-HQ     | 074ef77e62d3b29ae4042d7dfeb59c5ccbbe780516fc3f3a018724354f1df6d0 |
| B40, RTX 5090, RTX 5080, RTX PRO 6000 Blackwell Server Edition | 12.0 | 48k-HQ     | 6b3aaa8aa06da73326c853b4e9c5bf4f9634f63c594f84a943eaf330967a32b1 |
| B40, RTX 5090, RTX 5080, RTX PRO 6000 Blackwell Server Edition | 12.0 | 48k-LL     | 25321fe0e1b6fb55ba1413cef670bed81ec90e19817788de039fb30f26b0e054 |