Thursday, November 19, 2026 · No. 480Prices synced 19 hours ago
All·AI·Model

The AI Model Price Index

nemotron-3-ultra-nvfp4

by Requesty · family nemotron · listed Jun 2026RT

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.

Current rates USD per 1M tokens

Input$0.60
Output$2.40
Cache read$0.12
Cache write—

Against the market

Input price vs. maker flagships ($/1M tokens)
DeepSeek V4.1 Flash 1017DeepSeek V4.1 Flash 1017: $0.15 per 1M tokens$0.15Qwen3.8 Omni FlashQwen3.8 Omni Flash: $0.15 per 1M tokens$0.15DeepSeek V4.1 Flash 1017DeepSeek V4.1 Flash 1017: $0.30 per 1M tokens$0.30nemotron-3-ultra-nvfp4nemotron-3-ultra-nvfp4: $0.60 per 1M tokens$0.60Grok 4.7Grok 4.7: $2.00 per 1M tokens$2.00Gemini 4 NeonGemini 4 Neon: $3.00 per 1M tokens$3.00Claude Fable 5.2Claude Fable 5.2: $8.00 per 1M tokens$8.00GPT-6.2 AstraGPT-6.2 Astra: $8.00 per 1M tokens$8.00
Output price vs. maker flagships ($/1M tokens)
Qwen3.8 Omni FlashQwen3.8 Omni Flash: $0.47 per 1M tokens$0.47DeepSeek V4.1 Flash 1017DeepSeek V4.1 Flash 1017: $0.60 per 1M tokens$0.60DeepSeek V4.1 Flash 1017DeepSeek V4.1 Flash 1017: $1.20 per 1M tokens$1.20nemotron-3-ultra-nvfp4nemotron-3-ultra-nvfp4: $2.40 per 1M tokens$2.40Grok 4.7Grok 4.7: $6.00 per 1M tokens$6.00Gemini 4 NeonGemini 4 Neon: $15.00 per 1M tokens$15.00Claude Fable 5.2Claude Fable 5.2: $40.00 per 1M tokens$40.00GPT-6.2 AstraGPT-6.2 Astra: $40.00 per 1M tokens$40.00

On the record

Model id
nemotron-3-ultra-nvfp4
Context window
262K tokens
Max output
262K tokens
Input modalities
text
Output modalities
text
Reasoning
yes
Tool calls
yes
Open weights
no
Knowledge cutoff
—
Released
2026-06-13
Last updated
2026-06-13

Back of the envelope

≈

← All Requesty listings