Finance Agent Benchmark 2 — October 2026 evaluation Effort: High 42.6% vals.ai for Gemini 3 Flash (high) 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 Gemini 3 Flash on Finance Agent Benchmark 2 — October 2026 evaluation GPQA Diamond 90.4% blog.google for Gemini 3 Flash on Graduate-Level Google-Proof Q&A Diamond (opens in a new tab) GPQA Diamond Accuracy Reported December 17, 2025 Launch post Google Self-reported Google DeepMind Record for Gemini 3 Flash on Graduate-Level Google-Proof Q&A Diamond GPQA Diamond — Epoch AI harness Effort: High Configuration: Epoch AI harness 89.4% epoch.ai for Gemini 3 Flash (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 6, 2026 at 10:22 PM UTC Benchmark leaderboard Epoch AI Independent Record for Gemini 3 Flash on Graduate-Level Google-Proof Q&A Diamond — Epoch AI harness GPQA Diamond — Epoch AI harness Configuration: Epoch AI harness 83.2% epoch.ai for Gemini 3 Flash 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 December 17, 2025 at 10:39 PM UTC Benchmark leaderboard Epoch AI Independent Record for Gemini 3 Flash on Graduate-Level Google-Proof Q&A Diamond — Epoch AI harness GPQA Diamond — Vals AI harness Effort: High Configuration: Vals AI harness: mean of zero-shot and few-shot CoT 87.9% vals.ai for Gemini 3 Flash (high) 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 Gemini 3 Flash on Graduate-Level Google-Proof Q&A Diamond — Vals AI harness HLAB October 2026 evaluation Effort: High 0.0% vals.ai for Gemini 3 Flash (high) 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 Gemini 3 Flash on Harvey's Legal Agent Benchmark October 2026 evaluation Humanity's Last Exam Full set — No tools Configuration: No tools 33.7% blog.google for Gemini 3 Flash on Humanity's Last Exam Full set — No tools (opens in a new tab) Humanity's Last Exam Accuracy Reported December 17, 2025 Launch post Google Self-reported Google DeepMind Record for Gemini 3 Flash on Humanity's Last Exam Full set — No tools MCP Atlas Public, April 2026 update 62.0% labs.scale.com for Gemini 3 Flash 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 Gemini 3 Flash on MCP Atlas Public, April 2026 update MMLU-Pro Original — Vals AI harness Effort: High Configuration: Vals AI harness 88.6% vals.ai for Gemini 3 Flash (high) 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 Gemini 3 Flash on MMLU-Pro Original — Vals AI harness MMMU-Pro No tools Configuration: No tools 81.2% blog.google for Gemini 3 Flash on MMMU-Pro No tools (opens in a new tab) MMMU-Pro accuracy Setting: Overall (Standard 10 options + Vision) Reported December 17, 2025 Launch post Google Self-reported Google DeepMind Record for Gemini 3 Flash on MMMU-Pro No tools MMMU-Pro Vals AI harness Effort: High Configuration: Vals AI harness 87.6% vals.ai for Gemini 3 Flash (high) 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 Gemini 3 Flash on MMMU-Pro Vals AI harness SimpleQA Verified Original — Epoch AI harness Effort: High Configuration: Epoch AI harness 66.8% epoch.ai for Gemini 3 Flash (high) 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 7:30 PM UTC Benchmark leaderboard Epoch AI Independent Record for Gemini 3 Flash on SimpleQA Verified Original — Epoch AI harness SWE Atlas Codebase QnA — Public; mini-swe-agent Configuration: mini-swe-agent harness 8.2% labs.scale.com for Gemini 3 Flash on SWE Atlas Codebase QnA — Public; mini-swe-agent (opens in a new tab) Archived copy (opens in a new tab) Mean pass@1 Reported October 7, 2026 Benchmark leaderboard Scale AI Independent Record for Gemini 3 Flash on SWE Atlas Codebase QnA — Public; mini-swe-agent SWE Atlas Refactoring — mini-swe-agent Configuration: mini-swe-agent harness 10.0% labs.scale.com for Gemini 3 Flash on SWE Atlas Refactoring — mini-swe-agent (opens in a new tab) Archived copy (opens in a new tab) Mean pass@1 Reported October 7, 2026 Benchmark leaderboard Scale AI Independent Record for Gemini 3 Flash on SWE Atlas Refactoring — mini-swe-agent SWE-bench Multilingual — mini-swe-agent Configuration: mini-swe-agent harness 72.7% github.com for Gemini 3 Flash on SWE-bench Multilingual — mini-swe-agent (opens in a new tab) Archived copy (opens in a new tab) Resolved Tasks Reported February 13, 2026 Benchmark repository SWE-bench Independent SWE-bench Team Record for Gemini 3 Flash on SWE-bench Multilingual — mini-swe-agent SWE-bench Verified 78.0% blog.google for Gemini 3 Flash on SWE-bench Verified (opens in a new tab) Resolved Tasks Reported December 17, 2025 Launch post Google Self-reported Google DeepMind Record for Gemini 3 Flash on SWE-bench Verified SWE-bench Verified — Epoch AI harness Configuration: Epoch AI harness 75.4% epoch.ai for Gemini 3 Flash on SWE-bench Verified — Epoch AI harness (opens in a new tab) Archived copy (opens in a new tab) Resolved Tasks Reported October 7, 2026 Evaluated February 18, 2026 at 8:12 PM UTC Benchmark leaderboard Epoch AI Independent Record for Gemini 3 Flash on SWE-bench Verified — Epoch AI harness SWE-bench Verified — mini-swe-agent Effort: High Configuration: mini-swe-agent harness 75.8% github.com for Gemini 3 Flash (high) on SWE-bench Verified — mini-swe-agent (opens in a new tab) Archived copy (opens in a new tab) Resolved Tasks Reported February 17, 2026 Benchmark repository SWE-bench Independent SWE-bench Team Record for Gemini 3 Flash on SWE-bench Verified — mini-swe-agent SWE-bench Verified — Vals AI mini-swe-agent Effort: High Configuration: mini-swe-agent harness 75.0% vals.ai for Gemini 3 Flash (high) 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 Gemini 3 Flash on SWE-bench Verified — Vals AI mini-swe-agent SWE-bench Pro Public — Scale AI SWE-Agent Configuration: SWE-Agent harness, 250 turns, uncapped cost 34.6% labs.scale.com for Gemini 3 Flash on SWE-bench Pro Public — Scale AI SWE-Agent (opens in a new tab) Archived copy (opens in a new tab) Public tasks resolved Reported October 7, 2026 Benchmark leaderboard Scale AI Independent Record for Gemini 3 Flash on SWE-bench Pro Public — Scale AI SWE-Agent Terminal-Bench 2.0 — Vals AI Terminus 2 Effort: High Configuration: Terminus-2 harness 51.7% vals.ai for Gemini 3 Flash (high) on Terminal-Bench 2.0 — Vals AI Terminus 2 (opens in a new tab) Archived copy (opens in a new tab) Terminal-Bench Accuracy Reported June 4, 2026 Benchmark leaderboard Vals AI Independent Record for Gemini 3 Flash on Terminal-Bench 2.0 — Vals AI Terminus 2 Terminal-Bench 2.1 — Vals AI Terminus 2 Effort: High Configuration: Terminus-2 harness 53.9% vals.ai for Gemini 3 Flash (high) 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 Gemini 3 Flash on Terminal-Bench 2.1 — Vals AI Terminus 2 Toolathlon Original 49.4% toolathlon.xyz for Gemini 3 Flash on Toolathlon Original (opens in a new tab) Archived copy (opens in a new tab) Tasks completed Reported December 18, 2025 Benchmark leaderboard Toolathlon Independent Toolathlon Authors Record for Gemini 3 Flash on Toolathlon Original Vibe Code Bench 1.1 — OpenHands, October 2026 evaluation Effort: High Configuration: OpenHands harness 20.2% vals.ai for Gemini 3 Flash (high) 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 Gemini 3 Flash on Vibe Code Bench 1.1 — OpenHands, October 2026 evaluation τ³-bench τ³ — Airline Effort: High 82.5% sierra-tau-bench-public.s3.us-west-2.amazonaws.com for Gemini 3 Flash (high) on τ³-bench τ³ — Airline (opens in a new tab) Archived copy (opens in a new tab) Task success rate Reported March 2, 2026 Evaluated March 2, 2026 Benchmark leaderboard Sierra Independent Sierra Research Record for Gemini 3 Flash on τ³-bench τ³ — Airline τ³-bench τ³ — Banking, Terminal retrieval, v1.0.1 grading Effort: High Configuration: Terminal retrieval 27.3% sierra-tau-bench-public.s3.us-west-2.amazonaws.com for Gemini 3 Flash (high) on τ³-bench τ³ — Banking, Terminal retrieval, v1.0.1 grading (opens in a new tab) Archived copy (opens in a new tab) Task success rate Reported March 2, 2026 Evaluated March 2, 2026 Benchmark leaderboard Sierra Independent Sierra Research Record for Gemini 3 Flash on τ³-bench τ³ — Banking, Terminal retrieval, v1.0.1 grading τ³-bench τ³ — Retail Effort: High 76.8% sierra-tau-bench-public.s3.us-west-2.amazonaws.com for Gemini 3 Flash (high) on τ³-bench τ³ — Retail (opens in a new tab) Archived copy (opens in a new tab) Task success rate Reported March 2, 2026 Evaluated March 2, 2026 Benchmark leaderboard Sierra Independent Sierra Research Record for Gemini 3 Flash on τ³-bench τ³ — Retail τ³-bench τ³ — Telecom Effort: High 91.2% sierra-tau-bench-public.s3.us-west-2.amazonaws.com for Gemini 3 Flash (high) on τ³-bench τ³ — Telecom (opens in a new tab) Archived copy (opens in a new tab) Task success rate Reported March 2, 2026 Evaluated March 2, 2026 Benchmark leaderboard Sierra Independent Sierra Research Record for Gemini 3 Flash on τ³-bench τ³ — Telecom