APEX-Agents Original Effort: High 33.8% mercor.com for Inkling (high) 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 Inkling on APEX-Agents Original BrowseComp 130-question subset — Mercor web research agent Effort: High Configuration: Mercor web research agent, 100 steps, 3 hours 50.4% mercor.com for Inkling (high) on BrowseComp 130-question subset — Mercor web research agent (opens in a new tab) Archived copy (opens in a new tab) BrowseComp Accuracy Reported October 7, 2026 Benchmark leaderboard Mercor Independent Record for Inkling on BrowseComp 130-question subset — Mercor web research agent BrowseComp With context management — Inkling evaluation Reported as: effort=0.99 Configuration: Context management (Inkling evaluation) 77.1% thinkingmachines.ai for Inkling (effort=0.99) on BrowseComp With context management — Inkling evaluation (opens in a new tab) BrowseComp Accuracy Reported July 15, 2026 Launch post Thinking Machines Lab Self-reported Record for Inkling on BrowseComp With context management — Inkling evaluation Chartography October 2026 leaderboard Effort: High 10.0% surgehq.ai for Inkling (High) on Chartography October 2026 leaderboard (opens in a new tab) Archived copy (opens in a new tab) Chartography Pass@1 Reported October 7, 2026 Benchmark leaderboard Surge AI Independent Record for Inkling on Chartography October 2026 leaderboard CharXiv Descriptive and reasoning — Mercor single-shot Effort: High Configuration: Mercor single-shot agent 88.9% mercor.com for Inkling (high) on CharXiv Descriptive and reasoning — Mercor single-shot (opens in a new tab) Archived copy (opens in a new tab) CharXiv Accuracy Reported October 7, 2026 Benchmark leaderboard Mercor Independent Record for Inkling on CharXiv Descriptive and reasoning — Mercor single-shot CharXiv Reasoning — No tools Reported as: effort=0.99 Configuration: No tools 78.1% thinkingmachines.ai for Inkling (effort=0.99) on CharXiv Reasoning — No tools (opens in a new tab) CharXiv Accuracy Reported July 15, 2026 Model card Thinking Machines Lab Self-reported Record for Inkling on CharXiv Reasoning — No tools CharXiv Reasoning — Python enabled Reported as: effort=0.99 Configuration: With Python 82.0% thinkingmachines.ai for Inkling (effort=0.99) on CharXiv Reasoning — Python enabled (opens in a new tab) CharXiv Accuracy Reported July 15, 2026 Model card Thinking Machines Lab Self-reported Record for Inkling on CharXiv Reasoning — Python enabled CWE-bench 1 — OpenCode Effort: High Configuration: OpenCode harness 37.0% cwe-bench.com for Inkling (high) on CWE-bench 1 — OpenCode (opens in a new tab) Archived copy (opens in a new tab) CWE-bench Programmatic Pass@1 Reported October 7, 2026 Benchmark leaderboard Collinear AI Independent Record for Inkling on CWE-bench 1 — OpenCode DeepSWE 1.1 — Mercor mini-swe-agent, 500 steps Effort: High Configuration: Mercor mini-swe-agent, 500 steps, 2 hours 5.0% mercor.com for Inkling (high) 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 Inkling on DeepSWE 1.1 — Mercor mini-swe-agent, 500 steps Finance Agent Benchmark 2 — October 2026 evaluation Reported as: 0.99 46.6% vals.ai for Inkling (0.99) 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 Inkling on Finance Agent Benchmark 2 — October 2026 evaluation FrontierSWE 2 — October 2026 evaluation Configuration: Proximus harness, 5 trials, 20-hour budget 4.1% frontierswe.com for Inkling 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 Inkling on FrontierSWE 2 — October 2026 evaluation GPQA Diamond Reported as: effort=0.99 87.2% thinkingmachines.ai for Inkling (effort=0.99) on Graduate-Level Google-Proof Q&A Diamond (opens in a new tab) GPQA Diamond Accuracy Reported July 15, 2026 Model card Thinking Machines Lab Self-reported Record for Inkling on Graduate-Level Google-Proof Q&A Diamond GPQA Diamond — Epoch AI harness Effort: xHigh Configuration: Epoch AI harness 88.3% epoch.ai for Inkling (xhigh) 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 5, 2026 at 6:55 PM UTC Benchmark leaderboard Epoch AI Independent Record for Inkling on Graduate-Level Google-Proof Q&A Diamond — Epoch AI harness GPQA Diamond — Mercor single-shot Effort: High Configuration: Mercor single-shot agent 87.2% mercor.com for Inkling (high) 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 Inkling on Graduate-Level Google-Proof Q&A Diamond — Mercor single-shot GPQA Diamond — Vals AI harness Reported as: 0.99 Configuration: Vals AI harness: mean of zero-shot and few-shot CoT 87.1% vals.ai for Inkling (0.99) 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 Inkling on Graduate-Level Google-Proof Q&A Diamond — Vals AI harness HLAB October 2026 evaluation Reported as: 0.99 2.1% vals.ai for Inkling (0.99) 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 Inkling on Harvey's Legal Agent Benchmark October 2026 evaluation Humanity's Last Exam Text only — Mercor single-shot Effort: High Configuration: Mercor single-shot agent 25.5% mercor.com for Inkling (high) 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 Inkling on Humanity's Last Exam Text only — Mercor single-shot IFBench Original Reported as: effort=0.99 79.8% thinkingmachines.ai for Inkling (effort=0.99) on IFBench Original (opens in a new tab) Accuracy Reported July 15, 2026 Launch post Thinking Machines Lab Self-reported Record for Inkling on IFBench Original MCP Atlas Public, April 2026 update Effort: xHigh 76.0% labs.scale.com for Inkling (xHigh) 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 Inkling on MCP Atlas Public, April 2026 update MedXpertQA MM — Mercor single-shot Effort: High Configuration: Mercor single-shot agent 60.0% mercor.com for Inkling (high) on MedXpertQA MM — 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 Inkling on MedXpertQA MM — Mercor single-shot MMLU-Pro Original — Mercor single-shot Effort: High Configuration: Mercor single-shot agent 86.8% mercor.com for Inkling (high) 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 Inkling on MMLU-Pro Original — Mercor single-shot MMLU-Pro Original — Vals AI harness Reported as: 0.99 Configuration: Vals AI harness 86.3% vals.ai for Inkling (0.99) 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 Inkling on MMLU-Pro Original — Vals AI harness MMMU-Pro 1,730 questions — Mercor single-shot Effort: High Configuration: Mercor single-shot agent 69.7% mercor.com for Inkling (high) on MMMU-Pro 1,730 questions — Mercor single-shot (opens in a new tab) Archived copy (opens in a new tab) MMMU-Pro accuracy Setting: Not stated Reported October 7, 2026 Benchmark leaderboard Mercor Independent Record for Inkling on MMMU-Pro 1,730 questions — Mercor single-shot MMMU-Pro Vals AI harness Reported as: 0.99 Configuration: Vals AI harness 78.5% vals.ai for Inkling (0.99) on MMMU-Pro Vals AI harness (opens in a new tab) Archived copy (opens in a new tab) MMMU-Pro accuracy Setting: Standard (4 options) Reported September 1, 2026 Benchmark leaderboard Vals AI Independent Record for Inkling on MMMU-Pro Vals AI harness Riemann-bench October 2026 leaderboard Effort: xHigh 15.2% surgehq.ai for Inkling (xHigh) 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 Inkling on Riemann-bench October 2026 leaderboard SimpleQA Verified Original Reported as: effort=0.99 43.9% thinkingmachines.ai for Inkling (effort=0.99) on SimpleQA Verified Original (opens in a new tab) Accuracy Reported July 15, 2026 Launch post Thinking Machines Lab Self-reported Record for Inkling on SimpleQA Verified Original SimpleQA Verified Original — Epoch AI harness Effort: xHigh Configuration: Epoch AI harness 40.3% epoch.ai for Inkling (xhigh) 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 27, 2026 at 7:36 PM UTC Benchmark leaderboard Epoch AI Independent Record for Inkling on SimpleQA Verified Original — Epoch AI harness StrongREJECT Original Reported as: effort=0.99 98.6% thinkingmachines.ai for Inkling (effort=0.99) on StrongREJECT Original (opens in a new tab) Harmful-request refusal rate Reported July 15, 2026 Launch post Thinking Machines Lab Self-reported Record for Inkling on StrongREJECT Original SWE Atlas Codebase QnA — Mercor mini-swe-agent, 250 steps Effort: High Configuration: Mercor mini-swe-agent, 250 steps, 1 hour 21.2% mercor.com for Inkling (high) 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 Inkling on SWE Atlas Codebase QnA — Mercor mini-swe-agent, 250 steps SWE-bench Multilingual — Mercor mini-swe-agent, 1000 steps Effort: High Configuration: Mercor mini-swe-agent, 1000 steps, 3 hours 77.7% mercor.com for Inkling (high) 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 Inkling on SWE-bench Multilingual — Mercor mini-swe-agent, 1000 steps SWE-bench Verified — bash-only harness Reported as: effort=0.99 Configuration: bash-only harness 77.6% thinkingmachines.ai for Inkling (effort=0.99) on SWE-bench Verified — bash-only harness (opens in a new tab) Resolved Tasks Reported July 15, 2026 Launch post Thinking Machines Lab Self-reported Record for Inkling on SWE-bench Verified — bash-only harness SWE-bench Verified — Vals AI mini-swe-agent Reported as: 0.99 Configuration: mini-swe-agent harness 77.6% vals.ai for Inkling (0.99) 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 Inkling on SWE-bench Verified — Vals AI mini-swe-agent SWE-bench Pro Public Reported as: effort=0.99 54.3% thinkingmachines.ai for Inkling (effort=0.99) on SWE-bench Pro Public (opens in a new tab) Public tasks resolved Reported July 15, 2026 Launch post Thinking Machines Lab Self-reported Record for Inkling on SWE-bench Pro Public Terminal-Bench 2.1 — Thinking Machines internal coding harness Reported as: effort=0.99 Configuration: Thinking Machines internal coding harness 63.8% thinkingmachines.ai for Inkling (effort=0.99) on Terminal-Bench 2.1 — Thinking Machines internal coding harness (opens in a new tab) Terminal-Bench Accuracy Reported July 15, 2026 Launch post Thinking Machines Lab Self-reported Record for Inkling on Terminal-Bench 2.1 — Thinking Machines internal coding harness Terminal-Bench 2.1 — Vals AI Terminus 2 Reported as: 0.99 Configuration: Terminus-2 harness 47.6% vals.ai for Inkling (0.99) 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 Inkling on Terminal-Bench 2.1 — Vals AI Terminus 2 Terminal-Bench 4.0 — Mercor mini-swe-agent Effort: High Configuration: Mercor mini-swe-agent, 1000 steps, 8 hours 0.5% mercor.com for Inkling (high) 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 Inkling on Terminal-Bench 4.0 — Mercor mini-swe-agent Terminal-Bench 4.0 — Vals AI mini-swe-agent Reported as: 0.99 Configuration: mini-swe-agent harness 0.5% vals.ai for Inkling (0.99) 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 Inkling on Terminal-Bench 4.0 — Vals AI mini-swe-agent Toolathlon Verified Reported as: effort=0.99 45.5% thinkingmachines.ai for Inkling (effort=0.99) on Toolathlon Verified (opens in a new tab) Tasks completed Reported July 15, 2026 Launch post Thinking Machines Lab Self-reported Record for Inkling on Toolathlon Verified Vals Index 2.1 Reported as: 0.99 28.7% vals.ai for Inkling (0.99) 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 Inkling on Vals Index 2.1 Vibe Code Bench 1.1 — OpenHands, October 2026 evaluation Reported as: 0.99 Configuration: OpenHands harness 19.2% vals.ai for Inkling (0.99) 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 Inkling on Vibe Code Bench 1.1 — OpenHands, October 2026 evaluation τ³-bench τ³ — Banking, AllTools retrieval, v1.0.1 grading Effort: Max Configuration: AllTools retrieval 25.0% sierra-tau-bench-public.s3.us-west-2.amazonaws.com for Inkling (max) on τ³-bench τ³ — Banking, AllTools retrieval, v1.0.1 grading (opens in a new tab) Archived copy (opens in a new tab) Task success rate Reported August 4, 2026 Evaluated July 24, 2026 Benchmark leaderboard Sierra Independent Sierra Research Record for Inkling on τ³-bench τ³ — Banking, AllTools retrieval, v1.0.1 grading