OpenAI GPT-6 Astra (max)59.6% vals.ai for GPT-6 Astra (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 GPT-6 Astra on Terminal-Bench 4.0 — Vals AI mini-swe-agent 10005 OpenAI GPT-5.6 Luna (max)11.6% vals.ai for GPT-5.6 Luna (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 GPT-5.6 Luna on Terminal-Bench 4.0 — Vals AI mini-swe-agent 10009 OpenAI GPT-5.6 Sol (max)37.9% vals.ai for GPT-5.6 Sol (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 GPT-5.6 Sol on Terminal-Bench 4.0 — Vals AI mini-swe-agent 10010 OpenAI GPT-5.6 Terra (max)22.7% vals.ai for GPT-5.6 Terra (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 GPT-5.6 Terra on Terminal-Bench 4.0 — Vals AI mini-swe-agent 10011 OpenAI GPT-6 Luna (max)13.6% vals.ai for GPT-6 Luna (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 GPT-6 Luna on Terminal-Bench 4.0 — Vals AI mini-swe-agent 10013 OpenAI GPT-6 Sol (max)44.4% vals.ai for GPT-6 Sol (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 GPT-6 Sol on Terminal-Bench 4.0 — Vals AI mini-swe-agent 10014 OpenAI GPT-6.1 Sol (max)55.1% vals.ai for GPT-6.1 Sol (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 GPT-6.1 Sol on Terminal-Bench 4.0 — Vals AI mini-swe-agent 10015 OpenAI GPT-5.4 mini (xhigh)2.5% vals.ai for GPT-5.4 mini (xhigh) 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 GPT-5.4 mini on Terminal-Bench 4.0 — Vals AI mini-swe-agent 10021 DeepSeek AI DeepSeek-V4-Pro (max)11.1% vals.ai for DeepSeek-V4-Pro (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 DeepSeek-V4-Pro on Terminal-Bench 4.0 — Vals AI mini-swe-agent 110002 Moonshot AI Kimi K3 (max)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 120003 Alibaba (Tongyi) Qwen3.8-Max 34.3% vals.ai for Qwen3.8-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 Qwen3.8-Max on Terminal-Bench 4.0 — Vals AI mini-swe-agent 130008 MiniMax MiniMax M3 1.0% vals.ai for MiniMax M3 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 MiniMax M3 on Terminal-Bench 4.0 — Vals AI mini-swe-agent 140003 Z.ai GLM-5.3 (max)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 150005 Thinking Machines Lab Inkling (0.99)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 160001 Thinking Machines Lab Inkling-Small (0.99)0.0% vals.ai for Inkling-Small (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-Small on Terminal-Bench 4.0 — Vals AI mini-swe-agent 160002 Anthropic Claude Opus 4.8 23.2% vals.ai for Claude Opus 4.8 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 Claude Opus 4.8 on Terminal-Bench 4.0 — Vals AI mini-swe-agent 20011 Anthropic Claude Sonnet 5 9.6% vals.ai for Claude Sonnet 5 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 Claude Sonnet 5 on Terminal-Bench 4.0 — Vals AI mini-swe-agent 20013 Google (DeepMind) Gemini 3.8 Flash (high)19.2% vals.ai for Gemini 3.8 Flash (high) 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 Gemini 3.8 Flash on Terminal-Bench 4.0 — Vals AI mini-swe-agent 30004 Google (DeepMind) Gemini 3.1 Pro (high)2.5% vals.ai for Gemini 3.1 Pro (high) 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 Gemini 3.1 Pro on Terminal-Bench 4.0 — Vals AI mini-swe-agent 30005 Google (DeepMind) Gemini 3.5 Flash (high)6.1% vals.ai for Gemini 3.5 Flash (high) 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 Gemini 3.5 Flash on Terminal-Bench 4.0 — Vals AI mini-swe-agent 30006 Google (DeepMind) Gemini 3.7 Flash (high)12.1% vals.ai for Gemini 3.7 Flash (high) 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 Gemini 3.7 Flash on Terminal-Bench 4.0 — Vals AI mini-swe-agent 30010 Google (DeepMind) Gemini 4 Argon (high)57.6% vals.ai for Gemini 4 Argon (high) 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 Gemini 4 Argon on Terminal-Bench 4.0 — Vals AI mini-swe-agent 30011 SpaceXAI Grok 4.5 (high)8.6% vals.ai for Grok 4.5 (high) 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 Grok 4.5 on Terminal-Bench 4.0 — Vals AI mini-swe-agent 40001 SpaceXAI Grok 4.6 (high)17.2% vals.ai for Grok 4.6 (high) 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 Grok 4.6 on Terminal-Bench 4.0 — Vals AI mini-swe-agent 40002 SpaceXAI Grok 4.7 (xhigh)28.8% vals.ai for Grok 4.7 (xhigh) 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 Grok 4.7 on Terminal-Bench 4.0 — Vals AI mini-swe-agent 40003 Meta AI (originally Facebook AI Research) Muse Spark 1.2 (xhigh)6.1% vals.ai for Muse Spark 1.2 (xhigh) 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 Muse Spark 1.2 on Terminal-Bench 4.0 — Vals AI mini-swe-agent 80003 Meta AI (originally Facebook AI Research) Muse Spark 1.3 (xhigh)10.6% vals.ai for Muse Spark 1.3 (xhigh) 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 Muse Spark 1.3 on Terminal-Bench 4.0 — Vals AI mini-swe-agent 80005