NVIDIA

Nemotron 3 Nano Omni 30B A3B Reasoning

Nemotron 3 Nano Omni 30B A3B Reasoning is an open-weights 30B-parameter model by NVIDIA. It holds an Intelligence score of 34.1 on the local.ai leaderboard (2026-08 snapshot).

MoEreasoningopen weights
34.1
local.ai Intelligence
Parameters
30B
MoE · 3B active parameters
Benchmarks (local.ai snapshot)
τ²-bench
48%
GAIA
33%
GDPval
32%
HuggingFace metadata · updated 2026-05-08
repo
nvidia/Nemotron-3-Nano-Omni-30B-A3B-Reasoning-BF16
license
other
downloads
379,577
likes
414

Metadata refreshed from the HuggingFace API (snapshot 2026-08-15T15:53:42.405Z). Context length and long-form descriptions are the next pipeline stage.

Does it fit your hardware?

Detect or select your GPU first. Green = weights + context headroom, yellow = tight, grey = does not fit.

Q4_K_M 17.4GQ6_K 27GQ8_0 33GFP16 60G
Quantization≈ File sizeNote
IQ2_XXS8.1 GBExtreme compression, largest quality loss
IQ3_XXS10.8 GBVery small
Q3_K_M13.8 GBSmall
NVFP415 GBNVIDIA 4-bit
Q4_K_S16.2 GBCommon
Q4_K_M17.4 GBMost common sweet spot
Q5_K_M22.8 GBBalanced
Q6_K27 GBNear-lossless
Q8_033 GBNear-lossless
FP1660 GBOriginal weights

Actual files on HuggingFace

FileSize
model-00015-of-00017.safetensors3.72 GB
model-00006-of-00017.safetensors3.72 GB
model-00008-of-00017.safetensors3.72 GB
model-00010-of-00017.safetensors3.72 GB
model-00004-of-00017.safetensors3.72 GB
model-00002-of-00017.safetensors3.72 GB
GGUF: lmstudio-community/Nemotron-3-Nano-Omni-30B-A3B-Reasoning-GGUF · 62,148 downloads
FileSize
Nemotron-3-Nano-Omni-30B-A3B-Reasoning-Q4_K_M.gguf22.83 GBdownload
Nemotron-3-Nano-Omni-30B-A3B-Reasoning-Q6_K.gguf31.21 GBdownload
Nemotron-3-Nano-Omni-30B-A3B-Reasoning-Q8_0.gguf31.28 GBdownload
mmproj-Nemotron-3-Nano-Omni-30B-A3B-Reasoning-BF16.gguf1.48 GBdownload
GGUF: mudler/Nemotron-3-Nano-Omni-30B-A3B-Reasoning-APEX-GGUF · 14,737 downloads
FileSize
Nemotron-3-Nano-Omni-30B-A3B-Reasoning-APEX-Balanced.gguf25.34 GBdownload
Nemotron-3-Nano-Omni-30B-A3B-Reasoning-APEX-Compact.gguf18.94 GBdownload
Nemotron-3-Nano-Omni-30B-A3B-Reasoning-APEX-I-Balanced.gguf25.34 GBdownload
Nemotron-3-Nano-Omni-30B-A3B-Reasoning-APEX-I-Compact.gguf18.94 GBdownload
Nemotron-3-Nano-Omni-30B-A3B-Reasoning-APEX-I-Mini.gguf17.48 GBdownload
Nemotron-3-Nano-Omni-30B-A3B-Reasoning-APEX-I-Nano.gguf16.77 GBdownload
Nemotron-3-Nano-Omni-30B-A3B-Reasoning-APEX-I-Quality.gguf21.39 GBdownload
Nemotron-3-Nano-Omni-30B-A3B-Reasoning-APEX-Quality.gguf21.39 GBdownload
Nemotron-3-Nano-Omni-30B-A3B-Reasoning-F16.gguf58.84 GBdownload
mmproj.gguf1.48 GBdownload

What you save self-hosting this model

Uses the model's own API price when it exists; otherwise the closest commercial equivalent.

Waiting for hardware detection…

RunLocal presents third-party leaderboard data (local.ai / Exo Labs, 2026-08) with its own fit and cost estimates. Verify pricing, licenses and file sizes at the source before deploying a model.