Part of License & Price Change Watch — every vendor licence and price change we check, with the dates, and one free email the day a vendor changes terms.
Finds every NVIDIA NIM image (nvcr.io/nim), NGC key, NIM Helm chart and hosted NIM endpoint in Dockerfile, compose, Helm values and Terraform, and gives the SGLang or vLLM line that replaces each. Runs entirely in your browser — nothing is uploaded.
Same engine as the VS Code extension, byte for byte.
The whole workspace in one migration plan, a dated NIM licence exposure report (GPUs x $4,500 per year) and a CI gate that fails a build adding a new nvcr.io/nim pull.
One payment, one licence key for this tool. The key is shown right after payment.
Staying on NIM in production: NVIDIA AI Enterprise at $4,500 per GPU per year (about $1 per GPU per hour in the cloud), per NVIDIA's NIM FAQ
Buy the full version — $29It reads a Dockerfile, docker-compose.yml, Helm values.yaml or Terraform file and lists every NVIDIA NIM dependency: nvcr.io/nim images, NGC_API_KEY, the nvcr.io pull secret, the /opt/nim/.cache mount, NIM_* settings, the nim-llm chart, NIM Operator objects and hosted NIM endpoints. Each finding comes with the SGLang or vLLM line that replaces it, plus GPUs x $4,500 per year.
ML platform and DevOps engineers who prototyped an LLM service on NVIDIA NIM containers and are moving it to SGLang or vLLM before real users arrive, and engineering managers who need to know how many GPUs would fall under an NVIDIA AI Enterprise licence. It suits teams in the US, EU or anywhere NIM images run.
Grep for nvcr.io also hits plain CUDA base images such as nvcr.io/nvidia/cuda, and misses the NGC key, the NIM cache mount, NIM_* settings, Helm charts and hosted NIM endpoints. A chatbot cannot see your compose files. This tool gives file, line and the exact replacement line for each of 11 rules.
Free, with no key: the open file, every finding and every SGLang or vLLM replacement line. The full version is $29 once for one licence key per person or team seat: the whole workspace in one migration plan, a dated NIM licence exposure report and a CI gate that fails a build adding a new nvcr.io/nim pull.
NVIDIA's NIM FAQ says production use needs NVIDIA AI Enterprise, starting at $4,500 per GPU per year or about $1 per GPU per hour in the cloud, priced per GPU rather than per NIM. The bundled sample compose file reserves 3 GPUs, so 3 x $4,500 = $13,500 per year. SGLang and vLLM are open-source runtimes.
NVIDIA's FAQ ties free NIM access to Developer Program membership: development and testing on up to 16 GPUs, available for the duration of program membership. It names no end date. The change happens on your side, on the day the service starts serving real end users, which is production use under the FAQ.
One question, answered by the person who built it. Your email only if you want the answer sent.
Paste this code into your page. It runs in your visitor’s browser and stays up to date. Nothing your visitors type is stored.
For businesses: your name and logo instead of ours, and each visitor’s result + e-mail in your inbox (only when they tick consent) — Pro from $19/month. See Pro →