APEX-Agents Original Effort: Max 56.6% mercor.com for GLM-5.3 (max) on APEX-Agents Original (opens in a new tab) Archived copy (opens in a new tab) Pass rate Reported October 7, 2026 Benchmark leaderboard Mercor Independent Record for GLM-5.3 on APEX-Agents Original AutomationBench 1.0.6 Effort: Max 48.2% z.ai for GLM-5.3 (max) on AutomationBench 1.0.6 (opens in a new tab) Task pass rate Reported August 14, 2026 Launch post Z.ai Self-reported Record for GLM-5.3 on AutomationBench 1.0.6 CursorBench 4.0 Effort: Low 33.3% cursor.com for GLM-5.3 (low) on CursorBench 4.0 (opens in a new tab) Archived copy (opens in a new tab) CursorBench Accuracy Reported October 7, 2026 Benchmark leaderboard Cursor Independent Record for GLM-5.3 on CursorBench 4.0 CursorBench 4.0 Effort: High 38.0% cursor.com for GLM-5.3 (high) on CursorBench 4.0 (opens in a new tab) Archived copy (opens in a new tab) CursorBench Accuracy Reported October 7, 2026 Benchmark leaderboard Cursor Independent Record for GLM-5.3 on CursorBench 4.0 CursorBench 4.0 Effort: Max 42.6% cursor.com for GLM-5.3 (max) on CursorBench 4.0 (opens in a new tab) Archived copy (opens in a new tab) CursorBench Accuracy Reported October 7, 2026 Benchmark leaderboard Cursor Independent Record for GLM-5.3 on CursorBench 4.0 CyberGym Original Effort: Max 84.5% z.ai for GLM-5.3 (max) on CyberGym Original (opens in a new tab) Vulnerability reproduction success Reported August 14, 2026 Launch post Z.ai Self-reported Record for GLM-5.3 on CyberGym Original DeepSWE 1.1 Effort: Max 69.0% deepswe.datacurve.ai for GLM-5.3 (max) on DeepSWE 1.1 (opens in a new tab) DeepSWE Pass@1 Reported September 22, 2026 Benchmark leaderboard Datacurve Independent Record for GLM-5.3 on DeepSWE 1.1 DeepSWE 1.1 — Mercor mini-swe-agent, 500 steps Effort: Max Configuration: Mercor mini-swe-agent, 500 steps, 2 hours 70.5% mercor.com for GLM-5.3 (max) on DeepSWE 1.1 — Mercor mini-swe-agent, 500 steps (opens in a new tab) Archived copy (opens in a new tab) DeepSWE Pass@1 Reported October 7, 2026 Benchmark leaderboard Mercor Independent Record for GLM-5.3 on DeepSWE 1.1 — Mercor mini-swe-agent, 500 steps Finance Agent Benchmark 2 — October 2026 evaluation Effort: Max 55.8% vals.ai for GLM-5.3 (max) on Finance Agent Benchmark 2 — October 2026 evaluation (opens in a new tab) Archived copy (opens in a new tab) Finance Agent Accuracy Reported October 6, 2026 Benchmark leaderboard Vals AI Independent Record for GLM-5.3 on Finance Agent Benchmark 2 — October 2026 evaluation FrontierSWE 2 — October 2026 evaluation Configuration: Proximus harness, 5 trials, 20-hour budget 30.2% frontierswe.com for GLM-5.3 on FrontierSWE 2 — October 2026 evaluation (opens in a new tab) Archived copy (opens in a new tab) FrontierSWE Score Reported October 7, 2026 Benchmark leaderboard Proximal Independent Record for GLM-5.3 on FrontierSWE 2 — October 2026 evaluation GPQA Diamond — Epoch AI harness Effort: Max Configuration: Epoch AI harness 90.9% epoch.ai for GLM-5.3 (max) on Graduate-Level Google-Proof Q&A Diamond — Epoch AI harness (opens in a new tab) Archived copy (opens in a new tab) GPQA Diamond Accuracy Reported October 7, 2026 Evaluated August 24, 2026 at 7:44 PM UTC Benchmark leaderboard Epoch AI Independent Record for GLM-5.3 on Graduate-Level Google-Proof Q&A Diamond — Epoch AI harness GPQA Diamond — Mercor single-shot Effort: Max Configuration: Mercor single-shot agent 91.3% mercor.com for GLM-5.3 (max) on Graduate-Level Google-Proof Q&A Diamond — Mercor single-shot (opens in a new tab) Archived copy (opens in a new tab) GPQA Diamond Accuracy Reported October 7, 2026 Benchmark leaderboard Mercor Independent Record for GLM-5.3 on Graduate-Level Google-Proof Q&A Diamond — Mercor single-shot GPQA Diamond — Vals AI harness Effort: Max Configuration: Vals AI harness: mean of zero-shot and few-shot CoT 88.1% vals.ai for GLM-5.3 (max) on Graduate-Level Google-Proof Q&A Diamond — Vals AI harness (opens in a new tab) Archived copy (opens in a new tab) GPQA Diamond Accuracy Reported September 1, 2026 Benchmark leaderboard Vals AI Independent Record for GLM-5.3 on Graduate-Level Google-Proof Q&A Diamond — Vals AI harness HLAB October 2026 evaluation Effort: Max 8.3% vals.ai for GLM-5.3 (max) on Harvey's Legal Agent Benchmark October 2026 evaluation (opens in a new tab) Archived copy (opens in a new tab) HLAB Task Pass Rate Reported October 6, 2026 Benchmark leaderboard Vals AI Independent Record for GLM-5.3 on Harvey's Legal Agent Benchmark October 2026 evaluation Humanity's Last Exam Text only — Mercor single-shot Effort: Max Configuration: Mercor single-shot agent 40.0% mercor.com for GLM-5.3 (max) on Humanity's Last Exam Text only — Mercor single-shot (opens in a new tab) Archived copy (opens in a new tab) Humanity's Last Exam Accuracy Reported October 7, 2026 Benchmark leaderboard Mercor Independent Record for GLM-5.3 on Humanity's Last Exam Text only — Mercor single-shot MCP Atlas Public, April 2026 update 84.2% labs.scale.com for GLM-5.3 on MCP Atlas Public, April 2026 update (opens in a new tab) Pass rate Reported September 24, 2026 Benchmark leaderboard Scale AI Independent Record for GLM-5.3 on MCP Atlas Public, April 2026 update MMLU-Pro Original — Mercor single-shot Effort: Max Configuration: Mercor single-shot agent 88.0% mercor.com for GLM-5.3 (max) on MMLU-Pro Original — Mercor single-shot (opens in a new tab) Archived copy (opens in a new tab) Accuracy Reported October 7, 2026 Benchmark leaderboard Mercor Independent Record for GLM-5.3 on MMLU-Pro Original — Mercor single-shot MMLU-Pro Original — Vals AI harness Effort: Max Configuration: Vals AI harness 86.8% vals.ai for GLM-5.3 (max) on MMLU-Pro Original — Vals AI harness (opens in a new tab) Archived copy (opens in a new tab) Accuracy Reported September 1, 2026 Benchmark leaderboard Vals AI Independent Record for GLM-5.3 on MMLU-Pro Original — Vals AI harness NL2Repo Bench Original Effort: Max 58.0% z.ai for GLM-5.3 (max) on NL2Repo Bench Original (opens in a new tab) Average test pass rate Reported August 14, 2026 Launch post Z.ai Self-reported Record for GLM-5.3 on NL2Repo Bench Original PostTrainBench 1.2 — Claude Code Effort: Max Configuration: Claude Code harness 34.5% posttrainbench.com for GLM-5.3 (Max) on PostTrainBench 1.2 — Claude Code (opens in a new tab) Archived copy (opens in a new tab) PostTrainBench Weighted Aggregate Reported October 6, 2026 Benchmark leaderboard PostTrainBench Independent PostTrainBench Team Record for GLM-5.3 on PostTrainBench 1.2 — Claude Code Riemann-bench October 2026 leaderboard Effort: Max 12.0% surgehq.ai for GLM-5.3 (Max) on Riemann-bench October 2026 leaderboard (opens in a new tab) Archived copy (opens in a new tab) Riemann-bench Pass@1 Reported October 7, 2026 Benchmark leaderboard Surge AI Independent Record for GLM-5.3 on Riemann-bench October 2026 leaderboard SimpleQA Verified Original — Epoch AI harness Effort: Max Configuration: Epoch AI harness 41.0% epoch.ai for GLM-5.3 (max) on SimpleQA Verified Original — Epoch AI harness (opens in a new tab) Archived copy (opens in a new tab) Accuracy Reported October 7, 2026 Evaluated August 28, 2026 at 5:00 PM UTC Benchmark leaderboard Epoch AI Independent Record for GLM-5.3 on SimpleQA Verified Original — Epoch AI harness SWE Atlas Codebase QnA — Mercor mini-swe-agent, 250 steps Effort: Max Configuration: Mercor mini-swe-agent, 250 steps, 1 hour 42.5% mercor.com for GLM-5.3 (max) on SWE Atlas Codebase QnA — Mercor mini-swe-agent, 250 steps (opens in a new tab) Archived copy (opens in a new tab) Mean pass@1 Reported October 7, 2026 Benchmark leaderboard Mercor Independent Record for GLM-5.3 on SWE Atlas Codebase QnA — Mercor mini-swe-agent, 250 steps SWE-bench Multilingual — Mercor mini-swe-agent, 1000 steps Effort: Max Configuration: Mercor mini-swe-agent, 1000 steps, 3 hours 87.6% mercor.com for GLM-5.3 (max) on SWE-bench Multilingual — Mercor mini-swe-agent, 1000 steps (opens in a new tab) Archived copy (opens in a new tab) Resolved Tasks Reported October 7, 2026 Benchmark leaderboard Mercor Independent Record for GLM-5.3 on SWE-bench Multilingual — Mercor mini-swe-agent, 1000 steps SWE-bench Verified — Mercor mini-swe-agent, 1000 steps Effort: Max Configuration: Mercor mini-swe-agent, 1000 steps, 3 hours 81.7% mercor.com for GLM-5.3 (max) on SWE-bench Verified — Mercor mini-swe-agent, 1000 steps (opens in a new tab) Archived copy (opens in a new tab) Resolved Tasks Reported October 7, 2026 Benchmark leaderboard Mercor Independent Record for GLM-5.3 on SWE-bench Verified — Mercor mini-swe-agent, 1000 steps SWE-bench Verified — Vals AI mini-swe-agent Effort: Max Configuration: mini-swe-agent harness 95.4% vals.ai for GLM-5.3 (max) on SWE-bench Verified — Vals AI mini-swe-agent (opens in a new tab) Archived copy (opens in a new tab) Resolved Tasks Reported September 1, 2026 Benchmark leaderboard Vals AI Independent Record for GLM-5.3 on SWE-bench Verified — Vals AI mini-swe-agent Terminal-Bench 2.1 — Claude Code Effort: Max Configuration: Claude Code harness 88.2% z.ai for GLM-5.3 (max) on Terminal-Bench 2.1 — Claude Code (opens in a new tab) Terminal-Bench Accuracy Reported August 14, 2026 Launch post Z.ai Self-reported Record for GLM-5.3 on Terminal-Bench 2.1 — Claude Code Terminal-Bench 2.1 — Vals AI Terminus 2 Effort: Max Configuration: Terminus-2 harness 71.5% vals.ai for GLM-5.3 (max) on Terminal-Bench 2.1 — Vals AI Terminus 2 (opens in a new tab) Archived copy (opens in a new tab) Terminal-Bench Accuracy Reported September 28, 2026 Benchmark leaderboard Vals AI Independent Record for GLM-5.3 on Terminal-Bench 2.1 — Vals AI Terminus 2 Terminal-Bench 4.0 — Claude Code Effort: Max Configuration: Claude Code harness 41.8% tbench.ai for GLM-5.3 (max) on Terminal-Bench 4.0 — Claude Code (opens in a new tab) Archived copy (opens in a new tab) Terminal-Bench Accuracy Reported August 14, 2026 Benchmark leaderboard Terminal-Bench Independent Terminal-Bench Team Record for GLM-5.3 on Terminal-Bench 4.0 — Claude Code Terminal-Bench 4.0 — Mercor mini-swe-agent Effort: Max Configuration: Mercor mini-swe-agent, 1000 steps, 8 hours 37.9% mercor.com for GLM-5.3 (max) on Terminal-Bench 4.0 — Mercor mini-swe-agent (opens in a new tab) Archived copy (opens in a new tab) Terminal-Bench Accuracy Reported October 7, 2026 Benchmark leaderboard Mercor Independent Record for GLM-5.3 on Terminal-Bench 4.0 — Mercor mini-swe-agent Terminal-Bench 4.0 — Vals AI mini-swe-agent Effort: Max Configuration: mini-swe-agent harness 38.9% vals.ai for GLM-5.3 (max) on Terminal-Bench 4.0 — Vals AI mini-swe-agent (opens in a new tab) Archived copy (opens in a new tab) Terminal-Bench Accuracy Reported October 6, 2026 Benchmark leaderboard Vals AI Independent Record for GLM-5.3 on Terminal-Bench 4.0 — Vals AI mini-swe-agent Toolathlon Verified Effort: Max 73.0% z.ai for GLM-5.3 (max) on Toolathlon Verified (opens in a new tab) Tasks completed Reported August 14, 2026 Launch post Z.ai Self-reported Record for GLM-5.3 on Toolathlon Verified Vals Index 2.1 Effort: Max 53.5% vals.ai for GLM-5.3 (max) on Vals Index 2.1 (opens in a new tab) Archived copy (opens in a new tab) Vals Index GDP-weighted Accuracy Reported October 6, 2026 Benchmark leaderboard Vals AI Independent Record for GLM-5.3 on Vals Index 2.1 Vibe Code Bench 1.1 — OpenHands, October 2026 evaluation Effort: Max Configuration: OpenHands harness 78.1% vals.ai for GLM-5.3 (max) on Vibe Code Bench 1.1 — OpenHands, October 2026 evaluation (opens in a new tab) Archived copy (opens in a new tab) Vibe Code Bench Score Reported October 6, 2026 Benchmark leaderboard Vals AI Independent Record for GLM-5.3 on Vibe Code Bench 1.1 — OpenHands, October 2026 evaluation