Shared benchmarks

20
Shared benchmarks for GLM-5.1 and GLM-5.2
BenchmarkGLM-5.1GLM-5.2
  • 40.9%
  • 45.2%
Evaluation details for APEX-Agents Original

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

GLM-5.1

Record for GLM-5.1 on APEX-Agents Original

GLM-5.2

Record for GLM-5.2 on APEX-Agents Original
  • 6.2%
  • 26.2%
Evaluation details for AutomationBench 1.0.6

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

GLM-5.1

Record for GLM-5.1 on AutomationBench 1.0.6

GLM-5.2

Record for GLM-5.2 on AutomationBench 1.0.6
BrowseCompPotentially non-equivalent: Benchmark versions differ. Evaluator sets differ.
  • 68.0%BrowseComp · Z.ai
  • 65.4%130-question subset — Mercor web research agent · Mercor
Potentially non-equivalent · Evaluation details for BrowseComp

Benchmark versions differ. Evaluator sets differ. Evaluation methodology is not recorded by the registry.

GLM-5.1

Record for GLM-5.1 on BrowseComp BrowseComp

GLM-5.2

  • 68.7%
  • 77.2%
Evaluation details for CyberGym Original

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

GLM-5.1

Record for GLM-5.1 on CyberGym Original

GLM-5.2

Record for GLM-5.2 on CyberGym Original
  • 18.9%
  • 38.9%
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-5.1

Record for GLM-5.1 on DeepSWE 1.1 — Mercor mini-swe-agent, 500 steps

GLM-5.2

Record for GLM-5.2 on DeepSWE 1.1 — Mercor mini-swe-agent, 500 steps
  • 44.8%
  • 49.7%
Evaluation details for Finance Agent Benchmark 2 — October 2026 evaluation

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

  • 86.2%
  • 91.2%
Evaluation details for GPQA Diamond

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

GLM-5.1

Record for GLM-5.1 on Graduate-Level Google-Proof Q&A Diamond

GLM-5.2

Record for GLM-5.2 on Graduate-Level Google-Proof Q&A Diamond
  • 89.9%
  • 71.2%none
  • 87.9%low
  • 91.9%max
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-5.1

Record for GLM-5.1 on Graduate-Level Google-Proof Q&A Diamond — Epoch AI harness

GLM-5.2

Record for GLM-5.2 on Graduate-Level Google-Proof Q&A Diamond — Epoch AI harness
Record for GLM-5.2 on Graduate-Level Google-Proof Q&A Diamond — Epoch AI harness
Record for GLM-5.2 on Graduate-Level Google-Proof Q&A Diamond — Epoch AI harness
  • 84.5%
  • 85.6%
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-5.1

Record for GLM-5.1 on Graduate-Level Google-Proof Q&A Diamond — Vals AI harness

GLM-5.2

Record for GLM-5.2 on Graduate-Level Google-Proof Q&A Diamond — Vals AI harness
  • 0.0%
  • 7.1%
Evaluation details for HLAB October 2026 evaluation

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

GLM-5.1

Record for GLM-5.1 on Harvey's Legal Agent Benchmark October 2026 evaluation

GLM-5.2

Record for GLM-5.2 on Harvey's Legal Agent Benchmark October 2026 evaluation
  • 29.4%
  • 31.7%
Evaluation details for Humanity's Last Exam Text only — Mercor single-shot

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

GLM-5.1

Record for GLM-5.1 on Humanity's Last Exam Text only — Mercor single-shot

GLM-5.2

  • 75.6%
  • 77.8%
Evaluation details for MCP Atlas Public, April 2026 update

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

GLM-5.1

Record for GLM-5.1 on MCP Atlas Public, April 2026 update

GLM-5.2

Record for GLM-5.2 on MCP Atlas Public, April 2026 update
  • 86.9%
  • 86.7%
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-5.1

Record for GLM-5.1 on MMLU-Pro Original — Vals AI harness

GLM-5.2

Record for GLM-5.2 on MMLU-Pro Original — Vals AI harness
  • 42.7%
  • 48.9%
Evaluation details for NL2Repo Bench Original

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

GLM-5.1

Record for GLM-5.1 on NL2Repo Bench Original

GLM-5.2

Record for GLM-5.2 on NL2Repo Bench Original
  • 34.0%
  • 34.2%
Evaluation details for SimpleQA Verified Original — Epoch AI harness

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

  • 74.2%
  • 78.7%
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-5.1

Record for GLM-5.1 on SWE-bench Verified — Epoch AI harness

GLM-5.2

Record for GLM-5.2 on SWE-bench Verified — Epoch AI harness
  • 76.4%
  • 82.8%
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-5.1

Record for GLM-5.1 on SWE-bench Verified — Vals AI mini-swe-agent

GLM-5.2

Record for GLM-5.2 on SWE-bench Verified — Vals AI mini-swe-agent
  • 58.4%
  • 62.1%
Evaluation details for SWE-bench Pro Public

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

GLM-5.1

Record for GLM-5.1 on SWE-bench Pro Public

GLM-5.2

Record for GLM-5.2 on SWE-bench Pro Public
  • 56.9%
  • 67.8%
Evaluation details for Terminal-Bench 2.1 — Vals AI Terminus 2

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

GLM-5.1

Record for GLM-5.1 on Terminal-Bench 2.1 — Vals AI Terminus 2

GLM-5.2

Record for GLM-5.2 on Terminal-Bench 2.1 — Vals AI Terminus 2
  • 31.5%
  • 64.0%
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.