Ficelle
Nemotron 3 Super free API: where to use it for free
NVIDIA Nemotron 3 Super 120B A12B is free on Requesty and Cloudflare Workers AI. Exact ids, free-tier limits and failover.
Checked against NVIDIA's and each provider's official pages
Where Nemotron 3 Super is free
NVIDIA publishes the weights under its open model license, and two providers serve it on a free tier.
| Provider | Model id | Free-tier terms |
|---|---|---|
| Requesty | nvidia/nemotron-3-super-120b-a12b |
200 requests a day and 20 a minute for new organizations, shared across all free models. |
| Cloudflare | @cf/nvidia/nemotron-3-120b-a12b |
Within the shared 10,000 Neurons a day. 256,000-token context. |
Listed, but not free
-
OpenRouter
(
nvidia/nemotron-3-super-120b-a12b): paid only; the :free variant was withdrawn. -
NVIDIA
(
nvidia/nemotron-3-super-120b-a12b): trial credits for prototyping, not a free tier.
What Nemotron 3 Super is
Nemotron 3 Super is NVIDIA's open 120B mixture-of-experts model with 12B active parameters, a hybrid Mamba and attention design built for multi-agent workflows, long-context reasoning, tool use and RAG.
- Parameters
- 120B total, 12B active (MoE)
- Context window
- Up to 1M tokens
- Input
- Text
- Output
- Text
- Reasoning
- Switchable on or off
- Languages
- English, French, German, Italian, Japanese, Spanish, Chinese
Before you rely on it
- Hosted context is lower than the native 1M: about 256K on Cloudflare.
- NVIDIA has since released Nemotron 3 Ultra, also free on OpenRouter and Requesty.
Use Nemotron 3 Super with automatic failover
A single free tier runs out. Ficelle puts the free models of every provider you
configure behind one local OpenAI-compatible endpoint. Call a routing profile and
Nemotron 3 Super is one of the models it can pick; when it rate-limits or fails, the next
eligible free model answers. To pin Nemotron 3 Super alone, call the id that
ficelle models lists for it, knowing a pinned model has no failover.
from openai import OpenAI
client = OpenAI(base_url="http://127.0.0.1:8646/v1", api_key="ficelle-local")
# A routing profile keeps failover: Nemotron 3 Super is one of the free
# models it can pick, and the next eligible one answers if it fails.
client.chat.completions.create(
model="ficelle/auto-orchestrator",
messages=[{"role": "user", "content": "..."}],
)Routing profiles that fit what Nemotron 3 Super is built for:
- Agent orchestration
ficelle/auto-orchestrator - Tool calling
ficelle/auto-tools - Long context
ficelle/auto-long - Reasoning
ficelle/auto-reasoning
Whether a profile routes to Nemotron 3 Super on your install depends on your providers' live catalogs and Ficelle's eligibility checks, decided at request time.
Questions
Is Nemotron 3 Super free to use?
NVIDIA publishes the weights under its open model license, and two providers serve it on a free tier. When we checked on 28 September 2026, Requesty and Cloudflare served it at no charge, each under its own free-tier limits.
Where can I use the Nemotron 3 Super API for free?
Requesty as nvidia/nemotron-3-super-120b-a12b; Cloudflare as @cf/nvidia/nemotron-3-120b-a12b. Free catalogs change often, so check the provider's own page before you rely on one.
What is Nemotron 3 Super?
Nemotron 3 Super is NVIDIA's open 120B mixture-of-experts model with 12B active parameters, a hybrid Mamba and attention design built for multi-agent workflows, long-context reasoning, tool use and RAG.
Can I call Nemotron 3 Super through Ficelle?
Yes, when one of your configured providers serves it for free. Run ficelle models to see the ids your install can route to. Calling that id pins the one model, with no failover; calling a routing profile such as ficelle/auto-orchestrator keeps failover, so another free model answers when Nemotron 3 Super is rate-limited.
Sources
Official pages this guide rests on, read on 28 September 2026:
