GLM-5 glm-5

zai-org · glm hybrid-reasoning flagship agentic

#9 overall #3 agentic

params
754B
arch
moe
context
200k
license
mit open weights
released
Feb 2026
reasoning
yes
train compute
6.8e+24 FLOP

# architecture

attn shared · reasoning MHA ×64 router top-8 of 256 +1 shared expert ×256 expert out ×78 layers ctx 204,800 754B pool
layers
78
d_model
6144
heads
64q / 64kv
head dim
64
experts
top-8 of 256 + 1 shared
vocab
154880
family
glm_moe_dsa

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: #9 — score 86.3 ● high — all signals present

signalweightinput (0–100)
aa_intelligence 0.70 84.5
bench_composite 0.30 90.5

benchmark panel evidence:

complete panelscore (0–100)
browsecomp-v190.9
swe-bench-verified-v190.0

Coding: provisional — score 94.4 ● 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
aider_polyglot 0.30
swe_bench_verified 0.40 94.4

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

Agentic: #3 — score 94.4 ● low — a single signal; treat with caution

signalweightinput (0–100)
swe_bench_verified 0.60 94.4
swe_rebench 0.40

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

full methodology

# trend

OGM score · last 55 days
OGM score over 55 days
overall rank (up is better)
Overall rank over 55 days

methodology changed during this history window; score movement across that boundary is not model movement. See methodology v5.

# benchmarks

benchmarkscoresourcedate
AA Intelligence Index 32.4 Artificial Analysis
Browsecomp 0.8 / 1 LLM Stats
SWE-bench Verified 72.8 SWE-bench Feb 2026
LMArena Elo (Code) 1436.0 LMArena Aug 2026

# where to run

providerquantctx$/M in$/M out$/M cacheprice srctpsuptime
Featherless via hfrouter
GMICloudfp8 198k $0.60$1.92$0.12 via openrouter 99.6%
StreamLakefp8 198k $0.60$1.92$0.12 via openrouter 99.9%
DeepInfrafp4 198k $0.60$2.08$0.12 via openrouter 99.8%
Baidufp8 198k $0.70$2.24$0.14 via openrouter 100.0%
SiliconFlowfp8 200k $0.95$2.55$0.20 openrouter 99.9%
AtlasCloudfp8 198k $0.95$3.15 atlascloud
BaseTen 202k $0.95$3.15$0.20 via modelsdev
Amazon Bedrockunknown 198k $1.00$3.20 bedrock 95.9%
DigitalOceanunknown 64k $1.00$3.20$0.20 via openrouter 93.0%
Together 198k $1.00$3.20 via modelsdev
Venicefp8 198k $1.00$3.20 venice 85.2%
Z.AIfp8 198k $1.00$3.20$0.20 via openrouter 100.0%
Novitafp8 202k $1.20$4.00 novita 100.0%

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

# variants

GLM-5.1 GLM-5.2

source aliases
aa
glm-5
epoch
GLM-5
openrouter
z-ai/glm-5