This model requires you to accept the Apache 2.0 license terms before download.
Description
Granite Speech 3.3 8B — IBM speech-language model that pairs a speech encoder with the Granite 3.3 LLM for automatic speech recognition and speech-to-text translation across English, French, German, Spanish and Portuguese. Apache 2.0.
Tags
At a glance
| Kind | audio language model |
|---|---|
| Precision | bfloat16 |
| Container size | 16.1 GiB |
| Weights inside | 16.1 GiB |
| Components | 3 |
| Licence | Apache 2.0 |
| Origin | ibm-granite/granite-speech-3.3-8b @ 5670dcfae4a2 |
What is inside the container
A .nbx container packs every part of granite-speech-3.3-8b — 3 components with weights, graphs and runtime settings — so one file is all the runtime needs.
| Component | Blocks | Layers | Weights |
|---|---|---|---|
| encoder | 0 | — | 839 MiB |
| language_model | 0 | 40 | 15.2 GiB |
| projector | 0 | — | 72.1 MiB |
Default settings
The values the container ships with. They apply when you do not pass your own.
- Temperature
- 0
Memory needed
The weights of granite-speech-3.3-8b take 16.1 GiB in bfloat16. That is the floor for the memory the runtime has to place, spread over the devices it finds; activations come on top and grow with the size and length you ask for. A measured peak per device class is not published for this model yet.
Verify what you downloaded
Every download carries its SHA-256, and the file endpoint sends it in the X-NBX-SHA256 header. The container is 16.1 GiB (17,299,497,231 bytes).
$ sha256sum granite-speech-3.3-8b.nbx
b8b722adbf18d1569079ce69666fdb53b24b2a7f75494c16392691dd29bf246f
Questions
What is granite-speech-3.3-8b?
A audio language model by ibm-granite, distributed as a bfloat16 .nbx container and licensed Apache 2.0.
How big is the download?
16.1 GiB, of which 16.1 GiB are weights.
How do I run it?
Install the runtime from the download page, then use the commands under “CLI Usage” below: neurobrix import ibm-granite/granite-speech-3.3-8b, then neurobrix serve --model granite-speech-3.3-8b.
CLI Usage
$ neurobrix import ibm-granite/granite-speech-3.3-8b
$ neurobrix serve --model granite-speech-3.3-8b
$ neurobrix run --prompt "your prompt here"
$ neurobrix stop