NVIDIA Nemotron · Omni · Nemotron 3 Nano Omni
NVIDIA Nemotron 3 Nano Omni 30B-A3B Reasoning FP8 is a current Omni release in the NVIDIA Nemotron family, certified from its immutable official source revision.
NVIDIA Nemotron 3 Nano Omni 30B-A3B Reasoning FP8 is an official self-hostable NVIDIA release in the NVIDIA Nemotron Omni track. It uses NemotronH_Nano_Omni_Reasoning_V3, carries 31B / 3B MoE, supports a verified context window of 262,144 ټوکنونه, and exposes only source-certified checkpointونه and د اجرا چاپېریال combinations. Its release, license and deployment facts are tied to the official repository revision rather than inferred from the family name.
NVIDIA Nemotron 3 Nano Omni 30B-A3B Reasoning FP8 uses NemotronH_Nano_Omni_Reasoning_V3 with 31B / 3B MoE. The stored architecture profile records 64 layers, hidden width 8192, 64 attention heads and a 262,144 ټوکن context ceiling where the official configuration publishes those fields.
یوازې د سرچینې له خوا تصدیق شوې کره بڼې چې له دې نسخې سره تړلې دي، ښودل کېږي.
nvidia/Nemotron-3-Nano-Omni-30B-A3B-Reasoning-FP8
رسمي · FP8 · safetensors
اجرا چاپېریال تایید شوی: vLLM SGLang TensorRT-LLM Llama.cpp
| Precision / quantization | بڼه | بټونه | پر خپروونکي باور | د ځایپرځای کولو حالت |
|---|---|---|---|---|
| FP8 | safetensors | 8 | رسمي | اجرا چاپېریال تایید شوی |
Use only the listed vllm, sglang, tensorrt-llm, llama.cpp د اجرا چاپېریال configuration with the exact checkpoint precision. The recommendation engine checks weight memory, KV cache, د اجرا چاپېریال overhead, safety اضافي ظرفیت, GPU architecture and topology before ranking rentable and reference hardware.
د ارزښت له مخې غوره سپارښتنه
141 GB ټول VRAM · nvswitch
90.17 GB اضافي ظرفیت · محدود ظرفیت
USD ۲٬۳۴۵ / میاشتد عملي انتخابونو تر ټولو ټیټ لګښت
72 GB ټول VRAM · pcie
21.17 GB اضافي ظرفیت · شته
USD ۷۵۰ / میاشتتر ټولو لوړ فعالیت
141 GB ټول VRAM · nvswitch
90.17 GB اضافي ظرفیت · محدود ظرفیت
USD ۲٬۳۴۵ / میاشتThe full 31B resident weight set must fit across the selected GPUs. Context and concurrency add KV-cache memory; MoE active parameters do not replace resident-weight memory.
NVIDIA Nemotron 3 Nano Omni 30B-A3B Reasoning FP8 is published under NVIDIA. The release-scoped official license link and verification date are retained with the catalogue record.
سوداګریزه کارونه د شرطونو له مخې اجازه لري. سوداګریزه کارونه په شرطونو پورې تړلې ده. تر پلي کولو مخکې تړلي رسمي شرایط بیاکتئ.
د سرچینې revision: dd885e9548478d0d433c1239ed4598ad4a9faf7f58a8c706ab89885e5e2117fa
The exact estimate depends on checkpoint precision, context and concurrency; the page recommendation uses the certified memory engine and compatible د اجرا چاپېریال configurations.
Certified upstream د اجرا چاپېریال evidence is available for vllm, sglang, tensorrt-llm, llama.cpp.
The stored commercial classification is Commercial Use Allowed With Conditions. Consult the official NVIDIA terms for the final obligations.
وروستی تایید: 2026-08-26
