List your company

Long-horizon Task Solving on PINCHBENCH 23 tasks original

93.3Best Success RateClaude-Opus-4.6
43.27656.26369.2582.237Jun 15, 2026

Evaluation results

MethodLinks
Claude-Opus-4.6Source type=Closed-source models2026.0693.381.6
GPT-5.4Source type=Closed-source models2026.0690.579.4
MiniMax-M2.7Source type=Open-source / open-weight models2026.0689.882.8
Claude-Haiku-4.5Source type=Closed-source models2026.0689.577.4
MiMo-V2-FlashSource type=Open-source / open-weight models2026.0688.869.7
MiMo-V2-ProSource type=Closed-source models2026.0687.480.4
Gemini-3.1-Pro-PreviewSource type=Closed-source models2026.0686.777.5
OpenClaw-Skill 9BSource type=Open-source / open-weight models2026.0672.858.9
OpenClaw-Skill 4BSource type=Open-source / open-weight models2026.0671.556.4
Qwen3.5-4BSource type=Open-source / open-weight models2026.067155.7
Qwen3.5-9BSource type=Open-source / open-weight models2026.0667.553.8
GPT-OSS-120BSource type=Open-source / open-weight models2026.0667.152
GPT-OSS-20BSource type=Open-source / open-weight models2026.066650.3
OpenClaw-Skill-Qwen3 8BSource type=Open-source / open-weight models2026.0664.949.2
OpenClaw-Skill-Qwen3-4BSource type=Open-source / open-weight models2026.0664.547.9
Qwen3.6-PlusSource type=Open-source / open-weight models2026.0663.963.3
Qwen3-8BSource type=Open-source / open-weight models2026.0649.835.4
Llama 4 MaverickSource type=Open-source / open-weight models2026.0646.134.8
Qwen3-4BSource type=Open-source / open-weight models2026.0645.231.8