z-ai Open weights Reasoning

GLM 5.2 pricing & benchmarks

Released Jun 16, 2026 · served by 29 providers · ranked #13 by intelligence in our index

Input / 1M tokens$0.741
Output / 1M tokens$2.33
Context window1.0M
Cached input$0.138
Intelligence index51.1
Coding index68.8
Agentic index43.1
1M in + 300K out$1.44

What GLM 5.2 is

GLM 5.2 is a large-scale reasoning model from Z.ai. It supports text input and output with a 1M-token context window, and is suited for long-horizon agent workflows, project-level software engineering,...

Providers and prices for GLM 5.2

The same model costs different amounts depending on who serves it. Prices are per 1M tokens, cheapest first. Uptime and throughput are OpenRouter's rolling measurements, not vendor claims.

ProviderInputOutputContext QuantUptime 24hThroughput
StreamLakecheapest $0.741 $2.33 1.0M fp8 99.2%
Novita $0.742 $2.33 1.0M fp8 99.4%
CoreWeave $0.760 $2.42 262K fp4 96.9%
Baidu $0.770 $2.42 1.0M fp8 99.8%
AkashML $0.770 $2.42 97K fp8 96.6%
Decart $1.20 $2.50 1.0M fp8 94.8%
Inceptron $0.940 $2.90 1.0M fp4 93.3%
GMICloud $0.924 $2.90 1.0M fp8 96.2%
DeepInfra $0.930 $3.00 1.0M fp4 98.7%
Sail Research $1.12 $3.52 1.0M fp8 97.8%
Chutes $1.25 $3.95 1.0M fp4 94.6%
AtlasCloud $1.26 $3.96 1.0M fp8 99.2%
SiliconFlow $1.30 $4.09 1.0M fp8 99.2%
Morph $1.10 $4.10 1.0M 98.0%
DigitalOcean $1.05 $4.40 262K 98.4%
Ambient $1.05 $4.40 101K fp8 69.9%
Z.AI $1.40 $4.40 1.0M fp8 99.6%
Cloudflare $1.40 $4.40 262K 100.0%
Friendli $1.40 $4.40 1.0M 98.9%
Parasail $1.40 $4.40 262K fp4 90.4%
Venice $1.40 $4.40 1M fp8 98.0%
Together $1.40 $4.40 262K 98.3%
Ionstream $1.40 $4.40 1.0M fp4 82.2%
Phala $1.40 $4.40 1.0M 99.1%
Io Net $2.75 $6.02 262K fp8 99.5%
Wafer $2.10 $6.60 1.0M fp4 99.7%
Fireworks $2.10 $6.60 1.0M 96.4%
BaseTen $2.10 $6.60 524K fp8 99.9%
Alibaba $2.31 $7.26 1.0M 99.9%

GLM 5.2 benchmark results

Design Arena head-to-head results, as reported through the OpenRouter model API.

CategoryArenaRankEloWin rate
gamedev models #2 1358 62%
codecategories models #3 1346 60.8%
website models #3 1341 60.9%
3d models #4 1366 59.9%
fullstack agents #5 1279 63.8%
uicomponent models #5 1333 59.8%
webapps agents #7 1270 57%
htmlslides agents #9 1206 51.6%
mobileapps agents #9 1239 54.3%
svg models #9 1258 55.3%
androidnative agents #10 1237 54.8%
python-pptxslides agents #10 1185 48%
agenticgamedev agents #11 1187 48.4%
dataviz models #11 1299 56.7%
asciiart models #16 1219 48.8%
godotgamedev agents #17 1142 40.1%

Compare GLM 5.2 head-to-head

FAQ

How much does the GLM 5.2 API cost?

$0.741 per 1M input tokens and $2.33 per 1M output tokens. A typical workload of 1M input + 300K output tokens costs about $1.44.

Is GLM 5.2 free?

No. It is a paid model starting at $0.741 per 1M input tokens, though some providers offer trial credits.

Which provider is cheapest for GLM 5.2?

StreamLake at $0.741 input / $2.33 output per 1M tokens (fp8 quantization).

Can I self-host GLM 5.2?

Yes — weights are published as zai-org/GLM-5.2 on Hugging Face, so you can run it on your own hardware.