GLM-5.3

Z.ai
Arena score
1475
Among open models
#1of 148
Among all on Arena
#26
Votes
31,604

Measured on GLM-5.3 (Max), the served version, not a local build. LMArena, CC BY 4.0, published Nov 11, 2026.

Score by category

Top 10 among open models in#1Chinese#1Expert#1Hard prompts, English#1French

CategoryScore
  • Chinese1535
  • Expert1519
  • Applied math1502
  • Hard prompts, English1500
  • Math1500
  • Science1499
  • Coding1499
  • French1498
  • German1497
  • Software & IT1495
  • Hard prompts1493
  • Long prompts1486
  • Instructions1483
  • English1483
  • Medicine & health1483
  • Excluding ties1481
  • Legal & government1481
  • Multi-turn1474
  • Russian1472
  • Japanese1470
  • Writing & literature1468
  • Spanish1467
  • Polish1463
  • Non-English1462
  • Business & finance1461
  • Creative writing1459
  • Media & sports1447
  • Korean1440

Score by size

Plot against
Best for its size

About as strong, smaller

Compare

Discussion

1 post this week

Built on it

Quantizations
120
Fine-tunes
22
Adapters
0
Merges
2

Ways to run it

Most downloaded

BenchClub runs

Not runnable yet

Runs open once a GLM-5.3 build is registered.

More from Z.ai

See more

Other >300B models

See more