NIM endpoints by NVIDIA
Join the free NVIDIA Developer Program, create an API key on build.nvidia.com and call the OpenAI-compatible endpoint. Your current rate limit is shown in the top right of build.nvidia.com.
Quickstart
curl https://integrate.api.nvidia.com/v1/chat/completions \
-H "Authorization: Bearer $KEY" \
-d '{"model":"meta/llama-3.3-70b-instruct","messages":[{"role":"user","content":"hi"}]}'Using this model in production?
We wire free-tier and open-weight models into production stacks: evals against your real inputs, failover, cost caps. Book a free 30-minute call.
Frequently asked
Is NVIDIA NIM free?
Yes, for prototyping. NVIDIA's NIM FAQ says Developer Program members have free access to NIM API endpoints for prototyping, for as long as they stay members, and no credit card is needed. The old 1,000-credit allowance is gone: NVIDIA staff wrote on the developer forum in 2025 that the credit system was removed and trial use is governed by rate limits instead. Using NIM in production requires an NVIDIA AI Enterprise license.
Can I use NIM endpoints commercially?
The free tier is non-commercial. For production or commercial use you need a paid plan or a different model.
How do I start using NIM endpoints?
Join the free NVIDIA Developer Program, create an API key on build.nvidia.com and call the OpenAI-compatible endpoint. Your current rate limit is shown in the top right of build.nvidia.com.
What are the NVIDIA NIM free tier limits?
NVIDIA doesn't publish them. NVIDIA staff said on the developer forum in September 2025 that rate limits vary by model, aren't published, and that your maximum rate limit is shown in the top right of build.nvidia.com. Developers on the same forum commonly report about 40 requests per minute, but treat that as a community figure, not a promise.
Is NVIDIA NIM free forever?
There's no fixed end date. NVIDIA staff wrote in September 2025 that the trial period is not limited by a time period, and the NIM FAQ ties access to your Developer Program membership. It stays a prototyping tier that NVIDIA can change, so keep a paid or self-hosted fallback for production traffic.
Free monthly update
The monthly free-models update
What is newly free, what got rate-limited, and what to switch to. One email a month. Unsubscribe anytime.
Related free models
- GLM 4.6 (Z.ai) · Self-host free
- DeepSeek V3-0324 (DeepSeek) · Self-host free
- Gemini 2.5 Flash (Google AI Studio) · Free tier · existing users only, shuts down Dec 1, 2026
- Gemini 2.5 Pro (Google Gemini) · Free tier · existing users only, shuts down Dec 1, 2026
- Qwen3.7 Max (Qwen) · One-time 1M tokens, 90 days (new users)
- GPT OSS 120B (Scaleway) · First 1M tokens free (shared) · then paid