Reported by APEX-Agents
52.8%
Tasks passed · rank 19 of 39
- Unit
- % of tasks, higher is better
- Configuration
- GLM 5.3 Flash
- Measured
- 1 Oct 2026
- Not shown
- Not your firm's documents or tools; graded by rubric, not by a client.
Models / GLM 5.3 Flash
Z.ai
GLM 5.3 Flash
22 published results from 4 sources. Each card shows where the number comes from and what it does not measure. The overall leaderboard combines them; here each stands alone.
- Provider
- Z.ai
- Sources
- 4
- Our benchmarks
- 0
- Price per million tokens
- $0.15 in · $0.50 out
OpenRouter list price, 1 Oct 2026 · 1,048,576-token context
Reported by others
Reported by Arena (formerly LMArena)
0.07
Confirmed task success · rank 13 of 46
- Unit
- IPS effect estimate, higher is better
- Range
- 0.06 to 0.08
- Sample
- 55182 observations
- Configuration
- GLM 5.3 Flash
- Measured
- 28 Sep 2026
- Not shown
- Not a success rate: the estimated effect of choosing this model as the agent's orchestrator, against an even mix of the models tested, inside Arena's own harness.
Reported by Arena (formerly LMArena)
-0.00
Praise over complaint · rank 25 of 46
- Unit
- IPS effect estimate, higher is better
- Range
- -0.02 to 0.01
- Sample
- 22838 observations
- Configuration
- GLM 5.3 Flash
- Measured
- 28 Sep 2026
- Not shown
- Not a success rate: the estimated effect of choosing this model as the agent's orchestrator, against an even mix of the models tested, inside Arena's own harness.
Reported by Arena (formerly LMArena)
-0.01
Steerability · rank 29 of 46
- Unit
- IPS effect estimate, higher is better
- Range
- -0.02 to -0.00
- Sample
- 74818 observations
- Configuration
- GLM 5.3 Flash
- Measured
- 28 Sep 2026
- Not shown
- Not a success rate: the estimated effect of choosing this model as the agent's orchestrator, against an even mix of the models tested, inside Arena's own harness.
Reported by Arena (formerly LMArena)
0.00
Tool grounding · rank 13 of 46
- Unit
- IPS effect estimate, higher is better
- Range
- 0.00 to 0.00
- Sample
- 7664170 observations
- Configuration
- GLM 5.3 Flash
- Measured
- 28 Sep 2026
- Not shown
- Not a success rate: the estimated effect of choosing this model as the agent's orchestrator, against an even mix of the models tested, inside Arena's own harness.
Reported by Arena (formerly LMArena)
1,467
Overall · rank 33 of 177
- Unit
- Arena rating, higher is better
- Range
- 1,462 to 1,471
- Sample
- 18587 votes
- Configuration
- GLM 5.3 Flash
- Measured
- 25 Sep 2026
- Not shown
- Not an error rate: claims that cannot be checked on the web are skipped, and preference still carries most of the weight.
Reported by Arena (formerly LMArena)
1,463
Business, management and finance · rank 51 of 402
- Unit
- Arena rating, higher is better
- Range
- 1,453 to 1,474
- Sample
- 3682 votes
- Configuration
- GLM 5.3 Flash
- Measured
- 25 Sep 2026
- Not shown
- Not accuracy or correctness: it ranks which answer voters preferred, with answer length and formatting controlled for.
Reported by Arena (formerly LMArena)
1,433
Creative writing · rank 65 of 407
- Unit
- Arena rating, higher is better
- Range
- 1,423 to 1,444
- Sample
- 3992 votes
- Configuration
- GLM 5.3 Flash
- Measured
- 25 Sep 2026
- Not shown
- Not accuracy or correctness: it ranks which answer voters preferred, with answer length and formatting controlled for.
Reported by Arena (formerly LMArena)
1,510
Expert prompts · rank 25 of 359
- Unit
- Arena rating, higher is better
- Range
- 1,497 to 1,524
- Sample
- 2083 votes
- Configuration
- GLM 5.3 Flash
- Measured
- 25 Sep 2026
- Not shown
- Not accuracy or correctness: it ranks which answer voters preferred, with answer length and formatting controlled for.
Reported by Arena (formerly LMArena)
1,474
Instruction following · rank 26 of 409
- Unit
- Arena rating, higher is better
- Range
- 1,466 to 1,482
- Sample
- 6780 votes
- Configuration
- GLM 5.3 Flash
- Measured
- 25 Sep 2026
- Not shown
- Not accuracy or correctness: it ranks which answer voters preferred, with answer length and formatting controlled for.
Reported by Arena (formerly LMArena)
1,474
Overall · rank 35 of 409
- Unit
- Arena rating, higher is better
- Range
- 1,469 to 1,480
- Sample
- 19103 votes
- Configuration
- GLM 5.3 Flash
- Measured
- 25 Sep 2026
- Not shown
- Not accuracy or correctness: it ranks which answer voters preferred, with answer length and formatting controlled for.
Reported by Arena (formerly LMArena)
1,450
Writing, literature and language · rank 52 of 408
- Unit
- Arena rating, higher is better
- Range
- 1,441 to 1,459
- Sample
- 5215 votes
- Configuration
- GLM 5.3 Flash
- Measured
- 25 Sep 2026
- Not shown
- Not accuracy or correctness: it ranks which answer voters preferred, with answer length and formatting controlled for.
Reported by Artificial Analysis
71.5
Artificial Analysis Coding Index · rank 43 of 151
- Unit
- index score, higher is better
- Configuration
- GLM 5.3 Flash
- Measured
- 1 Oct 2026
- Not shown
- Not your codebase or tools.
Reported by Artificial Analysis
41.8
Artificial Analysis Intelligence Index · rank 47 of 314
- Unit
- index score, higher is better
- Configuration
- GLM 5.3 Flash
- Measured
- 1 Oct 2026
- Not shown
- Not business work, and a blend: read the parts for any one task.
Reported by Artificial Analysis
91.2%
GPQA Diamond · rank 49 of 276
- Unit
- % of questions, higher is better
- Configuration
- GLM 5.3 Flash
- Measured
- 1 Oct 2026
- Not shown
- Not applied work; multiple-choice science questions.
Reported by Artificial Analysis
39.9%
Humanity's Last Exam · rank 75 of 314
- Unit
- % of questions, higher is better
- Configuration
- GLM 5.3 Flash
- Measured
- 1 Oct 2026
- Not shown
- Not everyday work; academic questions at the edge of expertise.
Reported by Artificial Analysis
80.0%
Long-context reasoning (AA-LCR) · rank 65 of 309
- Unit
- % of questions, higher is better
- Configuration
- GLM 5.3 Flash
- Measured
- 1 Oct 2026
- Not shown
- Not retrieval over your own document store.
Reported by Artificial Analysis
51.6%
SciCode · rank 80 of 153
- Unit
- % of problems, higher is better
- Configuration
- GLM 5.3 Flash
- Measured
- 1 Oct 2026
- Not shown
- Not general software engineering.
Reported by Artificial Analysis
84.3%
Terminal-Bench 2.1 · rank 30 of 151
- Unit
- % of tasks, higher is better
- Configuration
- GLM 5.3 Flash
- Measured
- 1 Oct 2026
- Not shown
- Not other harnesses or tools; superseded by 4.0 for newer models.
Reported by Artificial Analysis
32.8%
Terminal-Bench 4.0 · rank 35 of 150
- Unit
- % of tasks, higher is better
- Configuration
- GLM 5.3 Flash
- Measured
- 1 Oct 2026
- Not shown
- Not other harnesses or tools; one attempt per task.
Reported by Artificial Analysis
47.2%
Τ-bench banking · rank 9 of 142
- Unit
- % of tasks, higher is better
- Configuration
- GLM 5.3 Flash
- Measured
- 1 Oct 2026
- Not shown
- Not your policies or systems; a simulated customer.
Reported by GDP.pdf
14.0%
Professional document tasks · rank 35 of 49
- Unit
- % of rubric, higher is better
- Configuration
- GLM 5.3 Flash (max)
- Measured
- 1 Oct 2026
- Not shown
- Not your documents; 100 tasks, so small differences are noise.