BrowseComp With context management — MiniMax M2.5 Configuration: Context management (MiniMax M2.5) 76.3% minimax.io for MiniMax M2.5 on BrowseComp With context management — MiniMax M2.5 (opens in a new tab) BrowseComp Accuracy Reported February 12, 2026 Launch post MiniMax Self-reported Record for MiniMax M2.5 on BrowseComp With context management — MiniMax M2.5 GPQA Diamond — Vals AI harness Configuration: Vals AI harness: mean of zero-shot and few-shot CoT 82.1% vals.ai for MiniMax M2.5 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 MiniMax M2.5 on Graduate-Level Google-Proof Q&A Diamond — Vals AI harness MMLU-Pro Original — Vals AI harness Configuration: Vals AI harness 80.1% vals.ai for MiniMax M2.5 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 MiniMax M2.5 on MMLU-Pro Original — Vals AI harness Multi-SWE-bench Original 51.3% minimax.io for MiniMax M2.5 on Multi-SWE-bench Original (opens in a new tab) Tasks resolved Reported February 12, 2026 Launch post MiniMax Self-reported Record for MiniMax M2.5 on Multi-SWE-bench Original SWE Atlas Codebase QnA — Public; mini-swe-agent Configuration: mini-swe-agent harness 10.3% labs.scale.com for MiniMax M2.5 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 MiniMax M2.5 on SWE Atlas Codebase QnA — Public; mini-swe-agent SWE Atlas Refactoring — mini-swe-agent Configuration: mini-swe-agent harness 19.5% labs.scale.com for MiniMax M2.5 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 MiniMax M2.5 on SWE Atlas Refactoring — mini-swe-agent SWE-bench Multilingual — mini-swe-agent Configuration: mini-swe-agent harness 68.3% github.com for MiniMax M2.5 on SWE-bench Multilingual — mini-swe-agent (opens in a new tab) Archived copy (opens in a new tab) Resolved Tasks Reported February 16, 2026 Benchmark repository SWE-bench Independent SWE-bench Team Record for MiniMax M2.5 on SWE-bench Multilingual — mini-swe-agent SWE-bench Verified 80.2% minimax.io for MiniMax M2.5 on SWE-bench Verified (opens in a new tab) Resolved Tasks Reported February 12, 2026 Launch post MiniMax Self-reported Record for MiniMax M2.5 on SWE-bench Verified SWE-bench Verified — Droid harness Configuration: Droid harness 79.7% minimax.io for MiniMax M2.5 on SWE-bench Verified — Droid harness (opens in a new tab) Resolved Tasks Reported February 12, 2026 Launch post MiniMax Self-reported Record for MiniMax M2.5 on SWE-bench Verified — Droid harness SWE-bench Verified — mini-swe-agent Effort: High Configuration: mini-swe-agent harness 75.8% github.com for MiniMax M2.5 (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 MiniMax M2.5 on SWE-bench Verified — mini-swe-agent SWE-bench Verified — OpenCode harness Configuration: OpenCode harness 76.1% minimax.io for MiniMax M2.5 on SWE-bench Verified — OpenCode harness (opens in a new tab) Resolved Tasks Reported February 12, 2026 Launch post MiniMax Self-reported Record for MiniMax M2.5 on SWE-bench Verified — OpenCode harness SWE-bench Verified — Vals AI mini-swe-agent Configuration: mini-swe-agent harness 74.2% vals.ai for MiniMax M2.5 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 MiniMax M2.5 on SWE-bench Verified — Vals AI mini-swe-agent Terminal-Bench 2.0 — Vals AI Terminus 2 Configuration: Terminus-2 harness 41.6% vals.ai for MiniMax M2.5 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 MiniMax M2.5 on Terminal-Bench 2.0 — Vals AI Terminus 2 Vibe Code Bench 1.1 — OpenHands, October 2026 evaluation Configuration: OpenHands harness 14.9% vals.ai for MiniMax M2.5 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 MiniMax M2.5 on Vibe Code Bench 1.1 — OpenHands, October 2026 evaluation