We are happy to announce we have expanded the LLamaMe service in the CZ and RZ with the release of v2.0 to support access to multiple on-prem large language models (LLMs). Existing LLamaMe API keys provisioned through LaunchIT now provide access to all of the following models within the applicable network zone.
For more information, see the LLamaMe documentation: https://hpc.llnl.gov/services/cloud-services/ai-ml-services/llamame-llm…
| Network | Models | GPUs |
|---|---|---|
| CZ |
meta-llama/Llama-3.3-70B-Instruct openai/gpt-oss-120b |
4 NVIDIA A100 80GB |
| RZ |
meta-llama/Meta-Llama-3.1-8B-Instruct
|
2 NVIDIA A100 40GB
|
|
Llama-3.3-70B-Instruct Codestral-22B-v0.1 Llama-4-Scout-17B-16E-Instruct gpt-oss-120b |
16 AMD MI250 120GB |