GLM-5.3 vs GPT-6 Sol

Score by job

1. Overall57n/a
2. Coding70~13
3. Agents~60n/a
4. Reasoning~41n/a
5. Math41~0
6. Writing6119
7. Design5468

Scores run from 0 to 100 within each job. A gap under 3 points is too close to call. ~ marks an early read from too few benchmarks to rank. How scores work

At a glance

Compare other models