Requesty
nemotron-3-ultra-nvfp4
Canonical ID:
requesty/nemotron-3-ultra-nvfp4
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.
Overview
Pricing and Limits
Token pricing
- Input
- $0.6/1M
- Output
- $2.4/1M
- Cached input
- N/A
Context and output
- Context window
- 262,144
- Max output tokens
- 262,144
- Qualified ID
- requesty/nemotron-3-ultra-nvfp4
Model Summary
nemotron-3-ultra-nvfp4 is a model listing in the Requesty provider catalog. It is categorized under the nemotron family.
The model supports text modalities across input/output paths. Reported context capacity is 262,144. Pricing is listed at $0.6/1M input and $2.4/1M output.
More From Requesty
Related models from the same provider catalog.
Requesty
Claude Fable 5
requesty/claude-fable-5
Requesty
Claude Fable 5.1
requesty/claude-fable-5.1
Requesty
Claude Fable 5.1 (EU)
requesty/claude-fable-5.1@eu
Requesty
Claude Fable 5 (EU)
requesty/claude-fable-5@eu
Requesty
Claude Haiku 4.5
requesty/anthropic/claude-haiku-4-5
Requesty
Claude Haiku 4.5 (latest)
requesty/claude-haiku-4-5
Requesty
Claude Haiku 4.5 (latest) (EU)
requesty/claude-haiku-4-5@eu
Requesty
Claude Haiku 5.5
requesty/claude-haiku-5-5
Same Family: nemotron
Comparable models in the same family across providers.
| Model | Provider | Context | Input | Output | Flags |
|---|---|---|---|---|---|
|
Llama 3.1 Nemotron 70B Instruct (Deep Infra)
edenai/deepinfra/nvidia/Llama-3.1-Nemotron-70B-Instruct
|
Eden AI
|
131,072 | $0.6/1M | $0.6/1M |
Tools
|
|
Llama 3.1 Nemotron Nano 8B v1
nvidia/nvidia/llama-3.1-nemotron-nano-8b-v1
|
Nvidia
|
131,072 | $0/1M | $0/1M |
Reasoning
Tools
|
|
Llama 3.1 Nemotron Nano VL 8B v1
nvidia/nvidia/llama-3.1-nemotron-nano-vl-8b-v1
|
Nvidia
|
32,768 | $0/1M | $0/1M |
Reasoning
Tools
|
|
llama-3.1-nemotron-safety-guard-8b-v3
nvidia/nvidia/llama-3.1-nemotron-safety-guard-8b-v3
|
Nvidia
|
128,000 | $0/1M | $0/1M |
|
|
Llama 3.3 Nemotron Super 49B v1.5
deepinfra/nvidia/Llama-3.3-Nemotron-Super-49B-v1.5
|
Deep Infra
|
131,072 | $0.4/1M | $0.4/1M |
Reasoning
Tools
|
|
Llama Nemotron Embed 1B v2
sap-ai-core/nvidia--llama-nemotron-embed-1b
|
SAP AI Core
|
8,192 | N/A | N/A |
|
|
llama-nemotron-embed-vl-1b-v2
nvidia/nvidia/llama-nemotron-embed-vl-1b-v2
|
Nvidia
|
32,768 | $0/1M | $0/1M |
|
|
llama-nemotron-rerank-vl-1b-v2
nvidia/nvidia/llama-nemotron-rerank-vl-1b-v2
|
Nvidia
|
128,000 | $0/1M | $0/1M |
|
|
mistral-nemotron
nvidia/mistralai/mistral-nemotron
|
Nvidia
|
128,000 | $0/1M | $0/1M |
Tools
|
|
Nemotron 3.5 Content Safety
nano-gpt/nvidia/nemotron-3.5-content-safety
|
NanoGPT
|
131,072 | $0.05/1M | $0.15/1M |
Reasoning
|