I have two sparks with cx7 interconnect. I’m trying to install Llama 405b exactly how nVidia did it. No other way seems to work. The problem is I’m unable to get access and download nvcr.io/nim/nvidia/llama-3_1-405b-instruct-fp4-spark:latest
Does anyone know how to get access? I have an API key but it doesn’t help with this specific container.
| Topic | Replies | Views | Activity | |
|---|---|---|---|---|
| Install Llama 405B on two Sparks with ConnectX-7 | 0 | 46 | April 20, 2026 | |
| NIM LLM Containers Fail on DGX Spark (GB10): Triton/vLLM Crash on sm_121 and NGC Permission Errors | 3 | 515 | February 6, 2026 | |
| DGX Spark / GB10 – nemotron-3-super-120b-a12b:1.8.0-variant fails downloading rl-030326-nvfp4 artifact (HTTP 400) | 20 | 1066 | March 26, 2026 | |
| Getting Started With NVIDIA NIM Tutorial Issues with NGC Registry | 6 | 2442 | July 24, 2024 | |
| Aunch NVIDIA NIM (llama3-8b-instruct) for LLMs locally | 3 | 257 | November 8, 2024 | |
| DGX Spark – Request AI Enterprise / NIM Entitlement Activation | 7 | 668 | January 6, 2026 | |
| Missing official native ARM64 NIM images for essential AI models | 6 | 587 | March 20, 2026 | |
| Running NIM llama-3_1-8b-instruct fails in On-Prem deployment | 7 | 638 | April 10, 2025 | |
| DGX Spark (GB10, ARM64) – Embedding NIM llama-3.2-nv-embedqa-1b-v2:1.10.0 fails with cudaErrorSymbolNotFound (onnx runtime) | 1 | 248 | December 17, 2025 | |
| Docker nvidia/vllm issue | 5 | 462 | March 12, 2026 |