GLM-4.7 vs GLM-5
10 shared benchmark contexts with reported scores and primary source links.
Change models in the comparison toolShared benchmarks
10| Benchmark | GLM-4.7 | GLM-5 |
|---|---|---|
|
| |
Evaluation details for BrowseCompThe recorded benchmark version, metric, and evaluator sets match. Evaluation methodology is not recorded by the registry. | ||
|
| |
Evaluation details for GPQA DiamondThe recorded benchmark version, metric, and evaluator sets match. Evaluation methodology is not recorded by the registry. | ||
|
| |
Evaluation details for GPQA Diamond — Epoch AI harnessThe recorded benchmark version, metric, and evaluator sets match. Evaluation methodology is not recorded by the registry. | ||
|
| |
Evaluation details for GPQA Diamond — Vals AI harnessThe recorded benchmark version, metric, and evaluator sets match. Evaluation methodology is not recorded by the registry. | ||
|
| |
Evaluation details for MMLU-Pro Original — Vals AI harnessThe recorded benchmark version, metric, and evaluator sets match. Evaluation methodology is not recorded by the registry. | ||
|
| |
Evaluation details for SWE-bench MultilingualThe recorded benchmark version, metric, and evaluator sets match. Evaluation methodology is not recorded by the registry. | ||
|
| |
Evaluation details for SWE-bench VerifiedThe recorded benchmark version, metric, and evaluator sets match. Evaluation methodology is not recorded by the registry. | ||
|
| |
Evaluation details for SWE-bench Verified — Vals AI mini-swe-agentThe recorded benchmark version, metric, and evaluator sets match. Evaluation methodology is not recorded by the registry. | ||
|
| |
Evaluation details for Terminal-Bench 2.0 — Vals AI Terminus 2The recorded benchmark version, metric, and evaluator sets match. Evaluation methodology is not recorded by the registry. | ||
|
| |
Evaluation details for Toolathlon OriginalThe recorded benchmark version, metric, and evaluator sets match. Evaluation methodology is not recorded by the registry. | ||