Llama-3.3-70b-Instruct

NVIDIA

Container

NVIDIA

Llama-3.3-70b-Instruct

This container houses the Llama-3.3-70B-Instruct, a 70B parameter, text-only language model for chat, coding and data generation. It offers strong performance comparable to larger models but with lower hardware requirements, and is optimized for dialogue.

NIM MetadataTo verify your workspace, run the Verify CLI locally and compare the generated hash with the NGC-published hash shown here. Learn more in the NVIDIA Documentation.Get detailed NIM model metadata, including properties, deployment, and configuration, via the NIM Metadata API.

			Engine		Actions
—	—	bf16	vllm	8	Copy NIM CLI Verification
h200	2335:10de	fp8	tensorrt_llm	1	Copy NIM CLI Verification
gb200	2941:10de	bf16	tensorrt_llm	1	Copy NIM CLI Verification
h200	2335:10de	fp8	tensorrt_llm	2	Copy NIM CLI Verification
gh200_480gb	2342:10de	fp8	tensorrt_llm	1	Copy NIM CLI Verification
a100_sxm4_40gb	20b0:10de	bf16	tensorrt_llm	8	Copy NIM CLI Verification
—	—	bf16	vllm	4	Copy NIM CLI Verification
h200_nvl	233b:10de	bf16	tensorrt_llm	1	Copy NIM CLI Verification
gh200_480gb	2342:10de	fp8	tensorrt_llm	1	Copy NIM CLI Verification
h200_nvl	233b:10de	bf16	tensorrt_llm	2	Copy NIM CLI Verification