Shared benchmarks

11
Shared benchmarks for GLM-5 and GLM-5.1
BenchmarkGLM-5GLM-5.1
  • 62.0%
  • 68.0%
Evaluation details for BrowseComp

The recorded benchmark version, metric, and evaluator sets match. Evaluation methodology is not recorded by the registry.

GLM-5

Record for GLM-5 on BrowseComp BrowseComp

GLM-5.1

Record for GLM-5.1 on BrowseComp BrowseComp
  • 43.2%
  • 68.7%
Evaluation details for CyberGym Original

The recorded benchmark version, metric, and evaluator sets match. Evaluation methodology is not recorded by the registry.

GLM-5

Record for GLM-5 on CyberGym Original

GLM-5.1

Record for GLM-5.1 on CyberGym Original
  • 2.1%
  • 18.9%
Evaluation details for DeepSWE 1.1 — Mercor mini-swe-agent, 500 steps

The recorded benchmark version, metric, and evaluator sets match. Evaluation methodology is not recorded by the registry.

GLM-5

Record for GLM-5 on DeepSWE 1.1 — Mercor mini-swe-agent, 500 steps

GLM-5.1

Record for GLM-5.1 on DeepSWE 1.1 — Mercor mini-swe-agent, 500 steps
  • 86.0%
  • 86.2%
Evaluation details for GPQA Diamond

The recorded benchmark version, metric, and evaluator sets match. Evaluation methodology is not recorded by the registry.

GLM-5

Record for GLM-5 on Graduate-Level Google-Proof Q&A Diamond

GLM-5.1

Record for GLM-5.1 on Graduate-Level Google-Proof Q&A Diamond
  • 87.8%
  • 89.9%
Evaluation details for GPQA Diamond — Epoch AI harness

The recorded benchmark version, metric, and evaluator sets match. Evaluation methodology is not recorded by the registry.

GLM-5

Record for GLM-5 on Graduate-Level Google-Proof Q&A Diamond — Epoch AI harness

GLM-5.1

Record for GLM-5.1 on Graduate-Level Google-Proof Q&A Diamond — Epoch AI harness
  • 83.3%
  • 84.5%
Evaluation details for GPQA Diamond — Vals AI harness

The recorded benchmark version, metric, and evaluator sets match. Evaluation methodology is not recorded by the registry.

GLM-5

Record for GLM-5 on Graduate-Level Google-Proof Q&A Diamond — Vals AI harness

GLM-5.1

Record for GLM-5.1 on Graduate-Level Google-Proof Q&A Diamond — Vals AI harness
  • 86.0%
  • 86.9%
Evaluation details for MMLU-Pro Original — Vals AI harness

The recorded benchmark version, metric, and evaluator sets match. Evaluation methodology is not recorded by the registry.

GLM-5

Record for GLM-5 on MMLU-Pro Original — Vals AI harness

GLM-5.1

Record for GLM-5.1 on MMLU-Pro Original — Vals AI harness
  • 72.1%
  • 74.2%
Evaluation details for SWE-bench Verified — Epoch AI harness

The recorded benchmark version, metric, and evaluator sets match. Evaluation methodology is not recorded by the registry.

GLM-5

Record for GLM-5 on SWE-bench Verified — Epoch AI harness

GLM-5.1

Record for GLM-5.1 on SWE-bench Verified — Epoch AI harness
  • 71.4%
  • 76.4%
Evaluation details for SWE-bench Verified — Vals AI mini-swe-agent

The recorded benchmark version, metric, and evaluator sets match. Evaluation methodology is not recorded by the registry.

GLM-5

Record for GLM-5 on SWE-bench Verified — Vals AI mini-swe-agent

GLM-5.1

Record for GLM-5.1 on SWE-bench Verified — Vals AI mini-swe-agent
  • 49.4%
  • 53.9%
Evaluation details for Terminal-Bench 2.0 — Vals AI Terminus 2

The recorded benchmark version, metric, and evaluator sets match. Evaluation methodology is not recorded by the registry.

GLM-5

Record for GLM-5 on Terminal-Bench 2.0 — Vals AI Terminus 2

GLM-5.1

Record for GLM-5.1 on Terminal-Bench 2.0 — Vals AI Terminus 2
  • 23.4%
  • 31.5%
Evaluation details for Vibe Code Bench 1.1 — OpenHands, October 2026 evaluation

The recorded benchmark version, metric, and evaluator sets match. Evaluation methodology is not recorded by the registry.