APEX-Agents Original Effort: Max 50.6% mercor.com for Kimi K3 (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 Kimi K3 on APEX-Agents Original AutomationBench 1.0.6 22.7% zapier.com for Kimi K3 on AutomationBench 1.0.6 (opens in a new tab) Archived copy (opens in a new tab) Task pass rate Reported October 7, 2026 Benchmark leaderboard Zapier Independent Zapier AutomationBench Team Record for Kimi K3 on AutomationBench 1.0.6 BabyVision Original — Python enabled Effort: Max Configuration: With Python 85.7% raw.githubusercontent.com for Kimi K3 (max) on BabyVision Original — Python enabled (opens in a new tab) Accuracy Reported August 6, 2026 Technical report Moonshot AI Self-reported Record for Kimi K3 on BabyVision Original — Python enabled BrowseComp 130-question subset — Mercor web research agent Effort: Max Configuration: Mercor web research agent, 100 steps, 3 hours 89.0% mercor.com for Kimi K3 (max) 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 Kimi K3 on BrowseComp 130-question subset — Mercor web research agent BrowseComp Effort: Max 91.2% huggingface.co for Kimi K3 (max) on BrowseComp BrowseComp (opens in a new tab) BrowseComp Accuracy Reported July 16, 2026 Model card Moonshot AI Self-reported Record for Kimi K3 on BrowseComp BrowseComp Chartography October 2026 leaderboard Effort: Max 26.6% surgehq.ai for Kimi K3 (Max) 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 Kimi K3 on Chartography October 2026 leaderboard CharXiv Descriptive and reasoning — Mercor single-shot Effort: Max Configuration: Mercor single-shot agent 93.9% mercor.com for Kimi K3 (max) 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 Kimi K3 on CharXiv Descriptive and reasoning — Mercor single-shot CharXiv Reasoning — No tools Effort: Max Configuration: No tools 84.8% huggingface.co for Kimi K3 (max) on CharXiv Reasoning — No tools (opens in a new tab) CharXiv Accuracy Reported July 16, 2026 Model card Moonshot AI Self-reported Record for Kimi K3 on CharXiv Reasoning — No tools CharXiv Reasoning — Python enabled Effort: Max Configuration: With Python 91.3% raw.githubusercontent.com for Kimi K3 (max) on CharXiv Reasoning — Python enabled (opens in a new tab) CharXiv Accuracy Reported August 6, 2026 Technical report Moonshot AI Self-reported Record for Kimi K3 on CharXiv Reasoning — Python enabled DeepSWE 1.1 Effort: Max 69.0% deepswe.datacurve.ai for Kimi K3 (max) on DeepSWE 1.1 (opens in a new tab) DeepSWE Pass@1 Reported September 22, 2026 Benchmark leaderboard Datacurve Independent Record for Kimi K3 on DeepSWE 1.1 DeepSWE 1.1 — Mercor mini-swe-agent, 500 steps Effort: Max Configuration: Mercor mini-swe-agent, 500 steps, 2 hours 66.4% mercor.com for Kimi K3 (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 Kimi K3 on DeepSWE 1.1 — Mercor mini-swe-agent, 500 steps Finance Agent Benchmark 2 — October 2026 evaluation Effort: Max 53.1% vals.ai for Kimi K3 (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 Kimi K3 on Finance Agent Benchmark 2 — October 2026 evaluation FrontierSWE 2 — October 2026 evaluation Configuration: Proximus harness, 5 trials, 20-hour budget 25.9% frontierswe.com for Kimi K3 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 Kimi K3 on FrontierSWE 2 — October 2026 evaluation GPQA Diamond Effort: Max 93.5% huggingface.co for Kimi K3 (max) on Graduate-Level Google-Proof Q&A Diamond (opens in a new tab) GPQA Diamond Accuracy Reported July 16, 2026 Model card Moonshot AI Self-reported Record for Kimi K3 on Graduate-Level Google-Proof Q&A Diamond GPQA Diamond — Epoch AI harness Effort: Low Configuration: Epoch AI harness 84.8% epoch.ai for Kimi K3 (low) 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 7, 2026 at 12:54 AM UTC Benchmark leaderboard Epoch AI Independent Record for Kimi K3 on Graduate-Level Google-Proof Q&A Diamond — Epoch AI harness GPQA Diamond — Epoch AI harness Effort: High Configuration: Epoch AI harness 91.9% epoch.ai for Kimi K3 (high) 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 7, 2026 at 12:53 AM UTC Benchmark leaderboard Epoch AI Independent Record for Kimi K3 on Graduate-Level Google-Proof Q&A Diamond — Epoch AI harness GPQA Diamond — Epoch AI harness Effort: Max Configuration: Epoch AI harness 93.1% epoch.ai for Kimi K3 (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 July 16, 2026 at 9:59 PM UTC Benchmark leaderboard Epoch AI Independent Record for Kimi K3 on Graduate-Level Google-Proof Q&A Diamond — Epoch AI harness GPQA Diamond — Mercor single-shot Effort: Max Configuration: Mercor single-shot agent 92.7% mercor.com for Kimi K3 (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 Kimi K3 on Graduate-Level Google-Proof Q&A Diamond — Mercor single-shot GPQA Diamond — Vals AI harness Configuration: Vals AI harness: mean of zero-shot and few-shot CoT 92.9% vals.ai for Kimi K3 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 Kimi K3 on Graduate-Level Google-Proof Q&A Diamond — Vals AI harness HLAB October 2026 evaluation Effort: Max 12.9% vals.ai for Kimi K3 (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 Kimi K3 on Harvey's Legal Agent Benchmark October 2026 evaluation Humanity's Last Exam Full set — No tools Effort: Max Configuration: No tools 43.5% huggingface.co for Kimi K3 (max) on Humanity's Last Exam Full set — No tools (opens in a new tab) Humanity's Last Exam Accuracy Reported July 16, 2026 Model card Moonshot AI Self-reported Record for Kimi K3 on Humanity's Last Exam Full set — No tools Humanity's Last Exam Full set — With tools Effort: Max Configuration: With tools 56.0% huggingface.co for Kimi K3 (max) on Humanity's Last Exam Full set — With tools (opens in a new tab) Humanity's Last Exam Accuracy Reported July 16, 2026 Model card Moonshot AI Self-reported Record for Kimi K3 on Humanity's Last Exam Full set — With tools Humanity's Last Exam Text only — Mercor single-shot Effort: Max Configuration: Mercor single-shot agent 46.5% mercor.com for Kimi K3 (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 Kimi K3 on Humanity's Last Exam Text only — Mercor single-shot MATH-Vision Original Effort: Max 94.3% raw.githubusercontent.com for Kimi K3 (max) on MATH-Vision Original (opens in a new tab) Accuracy Reported August 6, 2026 Technical report Moonshot AI Self-reported Record for Kimi K3 on MATH-Vision Original MATH-Vision Original — Python enabled Effort: Max Configuration: With Python 97.8% raw.githubusercontent.com for Kimi K3 (max) on MATH-Vision Original — Python enabled (opens in a new tab) Accuracy Reported August 6, 2026 Technical report Moonshot AI Self-reported Record for Kimi K3 on MATH-Vision Original — Python enabled MCP Atlas Public, April 2026 update Effort: Max 82.3% labs.scale.com for Kimi K3 (max) 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 Kimi K3 on MCP Atlas Public, April 2026 update MedXpertQA MM — Mercor single-shot Effort: Max Configuration: Mercor single-shot agent 73.2% mercor.com for Kimi K3 (max) 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 Kimi K3 on MedXpertQA MM — Mercor single-shot MMLU-Pro Original — Mercor single-shot Effort: Max Configuration: Mercor single-shot agent 89.3% mercor.com for Kimi K3 (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 Kimi K3 on MMLU-Pro Original — Mercor single-shot MMLU-Pro Original — Vals AI harness Effort: Max Configuration: Vals AI harness 88.0% vals.ai for Kimi K3 (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 Kimi K3 on MMLU-Pro Original — Vals AI harness MMMU-Pro 1,730 questions — Mercor single-shot Effort: Max Configuration: Mercor single-shot agent 80.7% mercor.com for Kimi K3 (max) 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 Kimi K3 on MMMU-Pro 1,730 questions — Mercor single-shot MMMU-Pro No tools Effort: Max Configuration: No tools 81.6% huggingface.co for Kimi K3 (max) on MMMU-Pro No tools (opens in a new tab) MMMU-Pro accuracy Setting: Not stated Reported July 16, 2026 Model card Moonshot AI Self-reported Record for Kimi K3 on MMMU-Pro No tools MMMU-Pro Vals AI harness Configuration: Vals AI harness 88.2% vals.ai for Kimi K3 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 Kimi K3 on MMMU-Pro Vals AI harness MMMU-Pro With tools Effort: Max Configuration: With tools 83.4% huggingface.co for Kimi K3 (max) on MMMU-Pro With tools (opens in a new tab) MMMU-Pro accuracy Setting: Not stated Reported July 16, 2026 Model card Moonshot AI Self-reported Record for Kimi K3 on MMMU-Pro With tools OSWorld Verified Effort: Max 84.8% huggingface.co for Kimi K3 (max) on OSWorld Verified (opens in a new tab) OSWorld-Verified success rate Reported July 16, 2026 Model card Moonshot AI Self-reported Record for Kimi K3 on OSWorld Verified PostTrainBench 1.1 — Claude Code Configuration: Claude Code harness 32.0% posttrainbench.com for Kimi K3 on PostTrainBench 1.1 — 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 Kimi K3 on PostTrainBench 1.1 — Claude Code PostTrainBench 1.2 — Claude Code Configuration: Claude Code harness 31.5% posttrainbench.com for Kimi K3 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 Kimi K3 on PostTrainBench 1.2 — Claude Code Riemann-bench October 2026 leaderboard Effort: Max 37.6% surgehq.ai for Kimi K3 (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 Kimi K3 on Riemann-bench October 2026 leaderboard SimpleQA Verified Original — Epoch AI harness Effort: Max Configuration: Epoch AI harness 50.6% epoch.ai for Kimi K3 (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 27, 2026 at 7:36 PM UTC Benchmark leaderboard Epoch AI Independent Record for Kimi K3 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 52.7% mercor.com for Kimi K3 (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 Kimi K3 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 91.4% mercor.com for Kimi K3 (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 Kimi K3 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 83.3% mercor.com for Kimi K3 (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 Kimi K3 on SWE-bench Verified — Mercor mini-swe-agent, 1000 steps SWE-bench Verified — Vals AI mini-swe-agent Configuration: mini-swe-agent harness 93.4% vals.ai for Kimi K3 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 Kimi K3 on SWE-bench Verified — Vals AI mini-swe-agent Terminal-Bench 2.1 Effort: Max 88.3% huggingface.co for Kimi K3 (max) on Terminal-Bench 2.1 (opens in a new tab) Terminal-Bench Accuracy Reported July 16, 2026 Model card Moonshot AI Self-reported Record for Kimi K3 on Terminal-Bench 2.1 Terminal-Bench 2.1 — Mercor Terminus 2 Effort: Max Configuration: Mercor Terminus 2, 1000 steps, 3 hours 82.0% mercor.com for Kimi K3 (max) on Terminal-Bench 2.1 — Mercor Terminus 2 (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 Kimi K3 on Terminal-Bench 2.1 — Mercor Terminus 2 Terminal-Bench 2.1 — Vals AI Terminus 2 Configuration: Terminus-2 harness 80.9% vals.ai for Kimi K3 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 Kimi K3 on Terminal-Bench 2.1 — Vals AI Terminus 2 Terminal-Bench 4.0 — Mercor mini-swe-agent Effort: Max Configuration: Mercor mini-swe-agent, 1000 steps, 8 hours 26.3% mercor.com for Kimi K3 (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 Kimi K3 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 17.2% vals.ai for Kimi K3 (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 Kimi K3 on Terminal-Bench 4.0 — Vals AI mini-swe-agent Toolathlon Verified Effort: Max 76.5% toolathlon.xyz for Kimi K3 (max) on Toolathlon Verified (opens in a new tab) Archived copy (opens in a new tab) Tasks completed Reported July 16, 2026 Benchmark leaderboard Toolathlon Independent Toolathlon Authors Record for Kimi K3 on Toolathlon Verified Vals Index 2.1 Effort: Max 50.3% vals.ai for Kimi K3 (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 Kimi K3 on Vals Index 2.1 Vibe Code Bench 1.1 — OpenHands, October 2026 evaluation Effort: Max Configuration: OpenHands harness 85.0% vals.ai for Kimi K3 (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 Kimi K3 on Vibe Code Bench 1.1 — OpenHands, October 2026 evaluation