Llama Models#
This page provides detailed technical specifications for the Llama model family supported by the NVIDIA NeMo Customizer microservice. For information about supported features and capabilities, refer to the Support Matrix in the Model Catalog.
Llama-3.3-70b Instruct#
Property |
Value |
|---|---|
Creator |
Meta |
Architecture |
transformer |
Description |
Llama-3.3-70b is a large language AI model optimized for advanced dialogue and reasoning capabilities. |
Max I/O Tokens |
8192 |
Parameters |
70 billion |
Training Data |
15+ trillion tokens (up to 2024) |
Recommended GPUs for Customization |
16 |
Default Name |
meta/llama-3.3-70b-instruct |
Version |
|
Llama-3.2-3b Instruct#
Property |
Value |
|---|---|
Creator |
Meta |
Architecture |
transformer |
Description |
Llama-3.2-3b is a compact yet powerful language model suitable for various dialogue applications. |
Max I/O Tokens |
8192 |
Parameters |
3 billion |
Training Data |
15+ trillion tokens (up to 2024) |
Recommended GPUs for Customization |
1 |
Default Name |
meta/llama-3.2-3b-instruct |
Version |
|
Llama-3.2-1b Instruct#
Property |
Value |
|---|---|
Creator |
Meta |
Architecture |
transformer |
Description |
Llama-3.2-1b is a lightweight language model designed for efficient deployment while maintaining strong capabilities. |
Max I/O Tokens |
8192 |
Parameters |
1 billion |
Training Data |
15+ trillion tokens (up to 2024) |
Recommended GPUs for Customization |
1 |
Default Name |
meta/llama-3.2-1b-instruct |
Version |
|
Llama-3.1-70b Instruct#
Property |
Value |
|---|---|
Creator |
Meta |
Architecture |
transformer |
Description |
Llama-3.1-70b is a large language AI model optimized for multilingual dialogue uses. |
Max I/O Tokens |
8192 |
Parameters |
70 billion |
Training Data |
15 trillion tokens (up to December 2023) |
Recommended GPUs for Customization |
16 |
Default Name |
meta/llama-3.1-70b-instruct |
Version |
|
Llama-3.1-8b Instruct#
Property |
Value |
|---|---|
Creator |
Meta |
Architecture |
transformer |
Description |
Llama-3.1-8b is a large language AI model optimized for multilingual dialogue uses. |
Max I/O Tokens |
8192 |
Parameters |
8 billion |
Training Data |
15 trillion tokens (up to December 2023) |
Recommended GPUs for Customization |
4 |
Default Name |
meta/llama-3.1-8b-instruct |
Version |
|
Llama-3-70b Instruct#
Property |
Value |
|---|---|
Creator |
Meta |
Architecture |
transformer |
Description |
Llama-3-70b is a large language AI model comprising a collection of models capable of generating text and code in response to prompts. |
Max I/O Tokens |
8192 |
Parameters |
70 billion |
Training Data |
15 trillion tokens (up to December 2023) |
Recommended GPUs for Customization |
16 |
Default Name |
meta/llama-3-70b-instruct |
Version |
|