Qwen2.5 72B Instruct qwen2.5-72b-instruct

qwen · qwen

#99 overall

params
72.7B
arch
dense
context
32k
license
qwen open weights
released
Sep 2024
downloads/30d
217.2k

# architecture

attn shared GQA 64q/8kv feed-forward (dense) every parameter, every token out ×80 layers ctx 32,768 72.7B — no routing
layers
80
d_model
8192
heads
64q / 8kv
head dim
128
vocab
152064
family
qwen2

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: #99 — score 10.2 ● low — a single signal; treat with caution

signalweightinput (0–100)
aa_intelligence 0.70 14.6
bench_composite 0.30 0.0

benchmark panel evidence:

complete panelscore (0–100)
aa-math-v10.0

full methodology

# trend

OGM score · last 106 days
OGM score over 106 days
overall rank (up is better)
Overall rank over 106 days

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

# benchmarks

benchmarkscoresourcedate
AA Intelligence Index 7.7 Artificial Analysis —
AA Math Index 14.0 Artificial Analysis —

# where to run

providerquantctx$/M in$/M out$/M cacheprice srctpsuptime✓
Featherless — ——— via hfrouter — —
SiliconFlow — ——— siliconflow — — ✓
Hyperbolic 128k $0.12$0.30— via litellm — —
Nebius 125k $0.13$0.40— via litellm — —
DeepInfrafp8 32k $0.36$0.40— deepinfra — 100.0% ✓
Novita 32k $0.38$0.40$0.38 via requesty — —
Together 32k $1.20$1.20— 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
qwen2.5-72b-instruct
arena
Qwen2.5-72B-Instruct
openrouter
qwen/qwen-2.5-72b-instruct