Route order log in sign up
1 pick your primary route

Z.ai: GLM 5.3

z-ai/glm-5.3 · agent · 1M ctx · added 2026-08-18
$0.1975·3.3287 /M effective hit 91% 797.47B/d

GLM-5.3 is a large-scale reasoning model from Z.ai, built for complex software engineering and long-horizon agent tasks. It supports text input and output with a 1M-token context window, and improves...

view
openrouter/z-ai/glm-5.3
Z.ai: GLM 5.3 · agent · 1M ctx
797.47B/d 91% $0.1975·3.3287 – 2mo
sticker
life 429 tok/d dp age
¢ ¢ᵉ ¢ⁱ ✓ ★
wafer/us
$0.0699·1.4000·0.0689 $0.0689·1.4000ⁱ $0.0690·1.3996ᵉ 90% – – 17% ✓ 616ms 90t/s 1d ·
wafer/fp4
$0.1790·1.4000·0.1780 $0.1780·1.4000ⁱ $0.1781·1.3996ᵉ 91% – – 3.2% ✓ 307ms 96t/s 1d ·
≈
decart/fp4
$0.5600·1.7700·0.1000 $0.1192·1.7700ⁱ $0.1430·1.7918ᵉ 91% – – 5.8% ✓ 656ms 83t/s 2mo ·
novita/fp8
$0.6300·1.9800·0.1170 $0.1384·1.9800ⁱ $0.1553·1.9798ᵉ 91% – – 3.0% ✓ 2004ms 70t/s 2mo ·
primeintellect
$0.4000·2.4640·0.0744 $0.0880·2.4640ⁱ $0.1139·2.4629ᵉ 86% – – 0.1% ✓ 1351ms 47t/s 1d ·
cloudflare
$0.7000·2.2000·0.1300 $0.1538·2.2000ⁱ $0.1561·2.1997ᵉ 95% – – 1.2% ✓ 3620ms 70t/s 2mo ·
siliconflow/fp8
$0.7000·2.2000·0.1300 $0.1538·2.2000ⁱ $0.1667·2.2003ᵉ 92% – – 1.5% ✓ 2528ms 53t/s 2mo ·
gmicloud/fp8
$0.7000·2.2000·0.1300 $0.1538·2.2000ⁱ $0.2074·2.1995ᵉ 87% – – 0.2% ◐ 3345ms 55t/s 2mo ·
reka
$0.1500·3.0000·0.1200 $0.1213·3.0000ⁱ $0.1306·3.0000ᵉ 57% – – 0.2% ✓ 878ms 46t/s 1d ·
deepinfra/fp4
$0.5625·2.5000·0.1250 $0.1433·2.5000ⁱ $0.1870·2.5254ᵉ 86% – – 1.3% ✓ 1497ms 64t/s 2mo ·
tenstorrent/fp8
$0.6000·2.5000·0.1000 $0.1209·2.5000ⁱ $0.1899·2.4997ᵉ 88% – – 0.8% ✓ 1902ms 104t/s ✕5h ·
inference-net/us
$0.0800·3.6000·0.0760 $0.0762·3.6000ⁱ $0.0762·3.5996ᵉ 95% – – 11% ✓ 312ms 109t/s 1d ·
inference-net/fp4
$0.0800·3.6000·0.0760 $0.0762·3.6000ⁱ $0.0762·3.5996ᵉ 94% – – 11% ✓ 359ms 117t/s 1d ·
phala
$0.8400·2.6400·0.1560 $0.1846·2.6400ⁱ $0.1896·2.6393ᵉ 94% – – <0.1% ✓ 940ms 58t/s ✕12h ·
sail-research/us
$0.7500·3.0000·0.1200 $0.1463·3.0000ⁱ $0.1611·3.0000ᵉ 92% – – 0.5% ✓ 1160ms 44t/s 1mo ·
≈
makora/fp4
$0.4000·3.2000·0.1600 $0.1700·3.2000ⁱ $0.1836·3.1996ᵉ 89% – – 0.7% ✓ 341ms 73t/s 2mo ·
relace/fp4
$0.0703·4.3843·0.0703 $0.0703·4.3843ⁱ $0.0703·4.3837ᵉ 88% – – 3.2% ✓ 1095ms 90t/s 1d ·
coreweave/nvfp4
$0.7500·3.0000·0.1900 $0.2134·3.0000ⁱ $0.2581·3.0000ᵉ 88% – – 3.6% ✓ 446ms 107t/s 1d ·
digitalocean
$0.9100·2.8600·0.1690 $0.2000·2.8600ⁱ $0.3201·2.8592ᵉ 78% – – 0.2% ✓ 617ms 102t/s 2mo ·
fireworks/spot
$0.9800·3.0800·0.1820 $0.2153·3.0800ⁱ $0.3102·3.0796ᵉ 86% – – 0.7% ✓ 792ms 94t/s 1d ·
streamlake/fp8
$0.7000·2.2000·0.1300 · $0.1538·2.2000ⁱ $0.3424·3.0225ᵉ · 80% · – – – · ◐ – · – · ✕2mo ·
akashml/fp4
$0.1900·4.4000·0.1900 $0.1900·4.4000ⁱ $0.1900·4.3998ᵉ 59% – – 0.3% ✓ 802ms 113t/s 1d ·
♻
river
$1.1200·3.5200·0.2080 $0.2461·3.5200ⁱ $0.2912·3.5198ᵉ 96% – – <0.1% ✓ 1611ms 29t/s ✕17h ·
≈
morph/fp8
$1.0530·2.1500·0.3560 $0.3851·2.1500ⁱ $0.4463·2.1497ᵉ 86% – – 0.3% ✓ 797ms 56t/s 1mo ·
baseten/fp4/us
$1.4000·4.4000·0.1400 $0.1926·4.4000ⁱ $0.2091·4.3995ᵉ 94% – – 1.3% ✓ 1747ms 64t/s 1d ·
alibaba
$1.1900·3.7400·0.2380 $0.2778·3.7400ⁱ $0.3153·3.7398ᵉ 92% – – 0.4% ◐ 1464ms 45t/s 1mo ·
inceptron/fp4
$1.0000·4.9000·0.1400 $0.1759·4.9000ⁱ $0.1990·4.8997ᵉ 92% – – 1.9% ✓ 657ms 102t/s 2mo ·
io-net/fp8
$1.0270·3.4760·0.2054 · $0.2397·3.4760ⁱ $0.3747·3.4754ᵉ · 79% · – – – · ✓ 1820ms · 64t/s · ✕1mo ·
baseten/fp4
$1.4000·4.4000·0.1400 $0.1926·4.4000ⁱ $0.2790·4.3996ᵉ 90% – – 0.2% ✓ 1487ms 61t/s 2mo ·
mistral/zdr
$1.4000·4.4000·0.1400 $0.1926·4.4000ⁱ $0.3173·4.3997ᵉ 87% – – 1.1% ✓ 1488ms 62t/s 1d ·
together
$1.4000·4.4000·0.2600 $0.3076·4.4000ⁱ $0.3229·4.3997ᵉ 93% – – 13% ✓ 972ms 83t/s 2mo ·
≈
baidu/fp8
$1.4000·4.4000·0.2600 $0.3076·4.4000ⁱ $0.3277·4.3995ᵉ 86% – – 0.3% ◐ – – 1mo ·
modal
$1.4000·4.4000·0.2600 $0.3076·4.4000ⁱ $0.3286·4.3996ᵉ 92% – – 1.7% ✓ 411ms 83t/s 2mo ·
sail-research/fp8
$1.0500·4.2000·0.1680 $0.2049·4.2000ⁱ $0.3525·4.1994ᵉ 80% – – <0.1% ✓ 454ms 73t/s 2mo ·
crusoe/fp4
$1.4000·4.4000·0.2600 $0.3076·4.4000ⁱ $0.3910·4.3997ᵉ 87% – – 0.6% ✓ 444ms 104t/s 2mo ·
fireworks
$1.4000·4.4000·0.2600 $0.3076·4.4000ⁱ $0.3954·4.3997ᵉ 86% – – 1.1% ✓ 1495ms 79t/s 2mo ·
mistral/nvfp4/eu
$1.5400·4.8400·0.1540 $0.2119·4.8400ⁱ $0.3479·4.8396ᵉ 87% – – 0.5% ✓ 983ms 54t/s 1d ·
parasail/fp8
$1.4000·4.4000·0.2600 $0.3076·4.4000ⁱ $0.4113·4.3996ᵉ 85% – – 0.7% ✓ 535ms 118t/s 2mo ·
z-ai/fp8
$1.4000·4.4000·0.2600 $0.3076·4.4000ⁱ $0.4553·4.3997ᵉ 86% – – 6.4% ✓ 1752ms 58t/s 2mo ·
friendli
$1.2600·3.9600·0.2340 $0.2769·3.9600ⁱ $0.5570·4.0175ᵉ 74% – – 0.1% ◐ 441ms 69t/s 2mo ·
nebius/fp4
$1.4000·4.4000·0.2600 $0.3076·4.4000ⁱ $0.5276·4.3992ᵉ 87% – – <0.1% ✓ 649ms 50t/s 1d ·
amazon-bedrock/us-east-1
$1.6800·5.2800·0.3120 $0.3692·5.2800ⁱ $0.4619·5.2795ᵉ 90% – – 1.4% ✓ 1237ms 136t/s 1d ·
atlas-cloud/fp8
$1.4000·4.4000·0.2600 $0.3076·4.4000ⁱ $0.5948·4.3994ᵉ 79% – – <0.1% ◐ 2294ms 48t/s 2mo ·
sambanova
$1.4000·4.4000·0.0700 $0.1256·4.4000ⁱ $0.6606·4.3991ᵉ 46% – – <0.1% ✓ 2487ms 58t/s 17h ·
venice
$1.4000·4.4000·0.2600 $0.3076·4.4000ⁱ $0.7534·4.4009ᵉ 62% – – <0.1% ✓ 818ms 85t/s 2mo ·
fireworks/fast
$2.1000·6.6000·0.3900 $0.4614·6.6000ⁱ $0.5383·6.5997ᵉ 92% – – 2.1% ✓ 1193ms 116t/s 1d ·
parasail/fast
$2.1000·6.6000·0.2600 $0.3369·6.6000ⁱ $0.6088·6.5998ᵉ 86% – – 0.4% ✓ 723ms 86t/s 1d ·
⚡
baseten/us/fast
$2.1000·6.6000·0.2100 $0.2890·6.6000ⁱ $0.6970·6.5998ᵉ 77% – – 0.2% ✓ 440ms 167t/s 1d ·
mistral/nvfp4
$1.4000·4.4000·0.1400 · $0.1926·4.4000ⁱ $0.9903·4.3999ᵉ · 33% · – – – · ◐ 2649ms · 60t/s · ✕1mo ·
baseten/fast
$2.1000·6.6000·0.2100 $0.2890·6.6000ⁱ $0.7710·6.5998ᵉ 72% – – 0.2% ✓ 547ms 148t/s 1d ·
alibaba/prime
$2.8000·8.8000·0.5600 $0.6536·8.8000ⁱ $0.7153·8.7997ᵉ 90% – – 0.2% ◐ 1054ms 75t/s 1d ·
reka/fp8
$0.8775·2.9700·0.1755 · $0.2048·2.9700ⁱ – · – · – – – · · 2039ms · 25t/s · ✕1mo ·
morph/fp4
$1.1200·3.5200·0.2240 · $0.2614·3.5200ⁱ – · – · – – – · · – · – · ✕2mo ·
io-net/fp4
$1.1250·4.3200·0.2610 · $0.2971·4.3200ⁱ – · – · – – – · · – · – · ✕2mo ·
akashml/fp8
$1.3000·4.4000·0.2600 · $0.3035·4.4000ⁱ – · – · – – – · · 532ms · 132t/s · ✕1mo ·
wafer
$1.4000·4.4000·0.2600 · $0.3076·4.4000ⁱ – · – · – – – · · 555ms · 123t/s · ✕1mo ·
baseten/fp8
$2.1000·6.6000·0.2100 · $0.2890·6.6000ⁱ – · – · – – – · · 30390ms · 7t/s · ✕1mo ·
vercel/zai/glm-5.3
GLM 5.3 · agent · 1M ctx
– 89% $0.3568·4.4000 12% 2mo
requesty/zai/glm-5.3
glm-5.3 · agent · text+image->text · 1M ctx
– 94% $0.3007·4.4000 1.0% 2mo
hf/zai-org/GLM-5.3
GLM-5.3 · agent · 1M ctx
– $1.4000·4.4000 – 2mo