GLM-4.6 glm-4.6

zai-org · glm hybrid-reasoning flagship agentic

#14 overall #3 coding #3 agentic

params
355B (A32B)
arch
moe
context
198k
license
mit open weights
released
Sep 2025
reasoning
yes
train compute
4.4e+24 FLOP

# why ranked

Overall: #14 — score 40.4 ● medium — one signal missing, weights renormalized

signalweightinput (0–100)
aa_intelligence 0.50 28.7
arena_elo 0.30
bench_composite 0.20 69.5

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

Coding: #3 — score 75.9 ● medium — one signal missing, weights renormalized

signalweightinput (0–100)
aa_coding 0.25 62.1
aider_polyglot 0.25
arena_elo_code 0.15
swe_bench_verified 0.35 85.7

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

Agentic: #3 — score 85.7 ● low — a single signal; treat with caution

signalweightinput (0–100)
swe_bench_verified 1.00 85.7

full methodology

# trend

collecting history — trends appear after a few daily runs.

# benchmarks

benchmarkscoresourcedate
AA Coding Index 45.8 Artificial Analysis
AA Intelligence Index 28.7 Artificial Analysis
AA Math Index 86.0 Artificial Analysis
AIME 2025 0.9 / 1 LLM Stats
Browsecomp 0.5 / 1 LLM Stats
SWE-bench Verified 68.2 SWE-bench Sep 2025

# where to run

providerquantctx$/M in$/M out$/M blendtpsuptime
DeepInfrafp4 198k $0.43$1.74$0.76 99.9%
Venicefp4 198k $0.43$1.75$0.76 100.0%
Novitabf16 200k $0.55$2.20$0.96 97.1%
AtlasCloudfp8 198k $0.60$2.20$1.00 100.0%
Z.AIfp4 198k $0.60$2.20$1.00 61.1%

blended = (3·input + output) / 4 per 1M tokens · cheapest first

# variants

GLM-4.6V

source aliases
aa
glm-4.6
arena
GLM-4.6
openrouter
z-ai/glm-4.6