Requesty
Sign in

identity

open weights
NVIDIA logo

nemotron-3-ultra-nvfp4NVIDIA

from /1M in$0.60fireworks ai
context262K262K out
providers11 region
endpoints11 region
endpoints
1
regions
1
api
chat
released
June 2026

capabilities 3/8

visionreasoningtoolscachingweb searchjson schemacomputer useimage gen

endpoints

1 / 1 region
Providers serving nemotron-3-ultra-nvfp4, with pricing and measured performance. Best value in each column is coloured.
#flags
1Fireworks AIGlobal262K$0.60$2.40$0.12zdr
provider price, no markupper 1M tokensblank where no qualifying sample

measured

through Sep 12

no measured traffic for this model yet. the endpoints table above carries provider pricing

benchmarks

artificial analysis

no published benchmark scores for this model

call it

model=
nemotron-3-ultra-nvfp4

requesty routes this id across every provider serving nemotron-3-ultra-nvfp4, picking on price and health and failing over automatically. the id stays valid when a provider changes underneath it

Base url is https://router.requesty.ai/v1 for every id here. One key reaches the whole catalog.

quickstart

openai compatible
main.py
from openai import OpenAI client = OpenAI(    api_key="YOUR_REQUESTY_API_KEY",    base_url="https://router.requesty.ai/v1",) response = client.chat.completions.create(    model="nemotron-3-ultra-nvfp4",    messages=[        {"role": "user", "content": "Explain quantum computing in one paragraph."},    ],) print(response.choices[0].message.content)

Change the base url, use your Requesty key, set the model to any id in the call pane. Existing OpenAI SDK code needs no other edit, and the same key reaches every other model in the catalog, so switching later is a one-parameter change. Browse all models

notes

reference

Nemotron-3-Ultra-550B-A55B-NVFP4 is a frontier-scale large language model (LLM) trained by NVIDIA, designed to deliver strong agentic, reasoning, and conversational capabilities. It is optimized for the most demanding workloads, including complex multi-step agents, long-context analysis, and high-accuracy reasoning over code, math, and science. The model employs a hybrid Latent Mixture-of-Experts (LatentMoE) architecture, utilizing interleaved Mamba-2 and MoE layers, along with select Attention layers. Like the Super model, the Ultra model incorporates Multi-Token Prediction (MTP) layers for faster text generation and improved quality, and it is trained using an NVFP4 pre-training recipe to maximize compute efficiency. The model has 55B active parameters and 550B parameters in total.

questions 6

Which providers serve nemotron-3-ultra-nvfp4?
nemotron-3-ultra-nvfp4 is available from 1 provider through Requesty: Fireworks AI. All endpoints share one OpenAI-compatible API and one key, and Requesty fails over between them automatically.
How much does nemotron-3-ultra-nvfp4 cost?
Pricing starts at $0.60 per million input tokens and $2.40 per million output tokens on the cheapest provider. Prices vary by provider and region; the table above shows every endpoint. Requesty charges exactly what the upstream provider charges, with no markup.
Is nemotron-3-ultra-nvfp4 open weights?
Yes. nemotron-3-ultra-nvfp4 is an open-weights model from NVIDIA, which is why multiple inference providers can host it. Provider choice affects price, latency, and data-privacy terms, all compared above.
What is the context window of nemotron-3-ultra-nvfp4?
nemotron-3-ultra-nvfp4 supports up to 262K tokens of context, with up to 262K output tokens per response. Some providers expose a smaller window; check the per-endpoint context column above.
How do I use nemotron-3-ultra-nvfp4 with the OpenAI SDK?
Install the OpenAI SDK, set base_url to "https://router.requesty.ai/v1", set your API key to your Requesty key, and set the model to "nemotron-3-ultra-nvfp4" or any endpoint id from the table above. The quickstart section shows Python, JavaScript and cURL snippets.
Should I call nemotron-3-ultra-nvfp4 by its managed id or a provider id?
Use the managed id, "nemotron-3-ultra-nvfp4". Requesty picks which of the 1 provider serves each request based on price and health, and fails over when one degrades, so the id keeps working while the routing changes underneath it. Send a provider id like "fireworks/nemotron-3-ultra-nvfp4" only when you need one specific deployment and want no failover.

more from nvidia 6

Other NVIDIA models
modelcontextfrom /Mproviders
nemotron-lightning-3.5-30b-a3b262K$0.051
nemotron-3.5-lightning-30b-a3b1.0Mfree1
nvidia-nemotron-3-ultra262K$0.371
nemotron-3.5-content-safety131Kfree1
nemotron-3-nano-omni-30b-a3b-reasoning131Kfree1
nemotron-3-ultra-550b-a55b1.0Mfree2

route nemotron-3-ultra-nvfp4 through one endpoint

One key for 1 provider on this model and 600+ others. No markup on provider prices, automatic failover, caching built in. Weekly aggregates in the measured pane come from production traffic routed through Requesty, one line per provider on a shared axis. Methodology

nemotron-3-ultra-nvfp41 provider1 endpointsfrom $0.60 /1M inctx 262Kupdated Sep 12