Nvidia Nemotron Nano 9B V2 nvidia-nemotron-nano-9b-v2

nvidia · nvidia

#52 overall

params
8.9B
arch
dense
license
other open weights
released
Aug 2025
downloads/30d
442.7k

# architecture

attn shared GQA 40q/8kv feed-forward (dense) every parameter, every token out ×56 layers 8.9B — no routing
layers
56
d_model
4480
heads
40q / 8kv
head dim
128
vocab
131072
family
nemotron_h

signal path of one layer, generated from the registry's structured fields — dimension lines quote the model's real numbers; an MoE trace forks at the router, a dense trace runs straight through. Geometry fields come from the repo's config.json.

# why ranked

Overall: #52 — score 15.8 ● high — all signals present

signalweightinput (0–100)
aa_intelligence 0.70 10.3
bench_composite 0.30 28.7

benchmark panel evidence:

complete panelscore (0–100)
aime-2025-v128.7

full methodology

# trend

OGM score · last 46 days
OGM score over 46 days
overall rank (up is better)
Overall rank over 46 days

# benchmarks

benchmarkscoresourcedate
AA Intelligence Index 3.2 Artificial Analysis
AA Math Index 69.7 Artificial Analysis
AIME 2025 0.7 / 1 LLM Stats
Bfcl-v3-multiturn 0.7 / 1 LLM Stats
Math-500 1.0 / 1 LLM Stats

# where to run

providerquantctx$/M in$/M out$/M cacheprice srctpsuptime
DeepInfra 128k $0.04$0.16 via litellm
Amazon Bedrock 125k $0.06$0.23 via litellm

sorted by blended price ((3·input + output) / 4 per 1M) · ✓ = the provider's own catalog confirms the offer · "via …" prices are what the aggregator routing the offer charges, not the provider's own list price

source aliases
aa
nvidia-nemotron-nano-9b-v2
llmstats
nvidia-nemotron-nano-9b-v2