Nex-N2-Pro nex-n2-pro

nex-agi · nex-agi

#13 overall

params
396.8B
arch
moe
license
apache-2.0 open weights
released
Jun 2026
downloads/30d
901

# architecture

attn shared GQA 32q/2kv router top-10 of 512 expert ×512 expert out ×60 layers 396.8B pool
layers
60
d_model
4096
heads
32q / 2kv
head dim
256
experts
top-10 of 512
vocab
248320
family
qwen3_5_moe_text

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: #13 — score 86.1 ● low — a single signal; treat with caution

signalweightinput (0–100)
aa_intelligence 0.70 84.6
bench_composite 0.30 89.6

benchmark panel evidence:

complete panelscore (0–100)
aa-coding-v189.6

Coding: provisional — score 79.4 ● low — a single signal; treat with caution

provisional: too few independent signals for a numbered position — the score is shown, but this model sorts after every ranked model.

signalweightinput (0–100)
aa_coding 0.30 79.4
aider_polyglot 0.30 —
swe_bench_verified 0.40 —

missing signals are dropped and the remaining weights renormalized — never imputed.

full methodology

# trend

OGM score · last 102 days
OGM score over 102 days
overall rank (up is better)
Overall rank over 102 days

methodology changed during this history window; score movement across that boundary is not model movement. See methodology v6.

# benchmarks

benchmarkscoresourcedate
AA Coding Index 59.1 Artificial Analysis —
AA Intelligence Index 28.2 Artificial Analysis —

# where to run

providerquantctx$/M in$/M out$/M cacheprice srctpsuptime✓
SiliconFlow 256k $0.50$2.50$0.25 via modelsdev — —

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
nex-n2-pro
openrouter
nex-agi/nex-n2-pro