AI Agent StoreNVIDIA · Newly covered hosted offer
Nemotron 3.5 Lightning · free Gateway offer
A newly listed free Gateway offer for NVIDIA’s August model. It supports text assistants, tools and long documents.
Released 2026-08-11 · First reviewed 2026-10-14 · Checked 2026-10-15.
We still need comparable independent tests. You can compare the confirmed features and limits below, but we cannot yet give this version an Agent Fit or Practical Value rating. Test it on your own tasks before switching; results for another version or thinking setting may not apply.
What you can use today
Token prices measure the small pieces of text the model reads or produces. Capacity is the amount the service accepts in a request; check the limits below for any fallback to another model. A dash means the rate is unverified.
Exact AI Gateway route ID: nvidia/nemotron-3.5-lightning-free. The hosted offer lists a 1M context, 65,000 output tokens and structured answers. The 30B downloadable model uses about 3B parameters per token; download hardware limits and licenses are separate.
Thinking settings listed by AI Gateway: none, low, medium, high. These controls apply to this AI Gateway service. We have not confirmed how they match the model maker’s settings, which is the default, or how they perform in comparable independent tests. Verify the exact Gateway catalog entry ↗
Gateway states that the provider may retain prompts and responses for model improvement. Consider that data policy before sending private work.
Gateway currently lists free input and output. Promotion duration, usage quotas and caching rates are unverified; this is not a promise of permanent free access. Your bill can also include tools, retries and charged thinking tokens. Check the offer used by your application.
Provider release and specifications ↗
Check availability and plans ↗
Test it before switching
Give each model the same examples, tools and definition of success. Record correct outcomes, response time, retries and total cost. A cheaper token price does not guarantee a cheaper completed task.
Compare caching and long-context costs ↗