model_name,grounding,instruction_following,planning,reasoning,refinement,safety,theory_of_mind,tool_usage,multilingual phi-1,1.0375,1.01,1.0,1.0,1.0172413793103448,1.3768115942028984,1.0,1.0125, phi-1_5,2.45,2.84,2.257142857142857,2.12,2.1724137931034484,2.9130434782608696,2.62,1.275, phi-2,2.9625,2.75,2.7142857142857144,2.69,2.5689655172413794,3.4347826086956523,2.98,1.65, Qwen1.5-0.5B,1.925,2.04,1.6,1.51,1.5,1.9565217391304348,1.72,1.1875, Qwen1.5-1.8B,2.425,2.7,2.2285714285714286,1.81,2.086206896551724,2.449275362318841,2.38,1.35, Qwen1.5-4B,2.7875,2.89,2.442857142857143,2.23,2.1551724137931036,3.2753623188405796,2.51,1.675, gemma-2b,2.25,2.65,2.085714285714286,1.94,1.8620689655172413,2.63768115942029,2.31,1.2875, OLMo-1B,1.675,1.64,1.3571428571428572,1.31,1.3103448275862069,2.0869565217391304,1.44,1.0625, Qwen1.5-0.5B-Chat,2.075,2.44,1.9142857142857144,1.64,1.6896551724137931,2.420289855072464,2.26,1.25,1.1857142857142857 Qwen1.5-1.8B-Chat,2.85,3.11,2.642857142857143,2.24,2.5172413793103448,2.7246376811594204,3.11,1.6625,1.3285714285714285 Qwen1.5-4B-Chat,2.8,3.1,2.8714285714285714,2.53,2.8620689655172415,3.347826086956522,3.0,1.9375,1.4714285714285715 Phi-3-mini-4k-instruct,3.9,3.85,3.4857142857142858,3.54,3.7758620689655173,4.231884057971015,3.81,3.0625,1.9714285714285715 Phi-3-mini-128k-instruct,3.5875,3.66,3.4714285714285715,3.66,3.3448275862068964,3.9420289855072466,3.7,2.9125,1.8142857142857143 gemma-2b-it,2.8,3.09,2.9714285714285715,2.36,2.6379310344827585,4.043478260869565,3.12,1.75,1.6857142857142857 gemma-1.1-2b-it,2.9125,3.29,3.0285714285714285,2.55,2.706896551724138,4.130434782608695,3.25,1.675,1.6571428571428573 gemma-7b,1.375,1.46,1.2142857142857142,1.22,1.0344827586206897,1.9275362318840579,1.19,1.0125, Mistral-7B-v0.1,2.9375,3.23,2.914285714285714,2.68,2.4655172413793105,3.4057971014492754,2.9,1.975, Mistral-7B-v0.2,3.025,3.24,2.7857142857142856,2.58,2.4827586206896552,3.2028985507246377,3.07,1.8625, Qwen1.5-7B,2.9375,3.0,2.842857142857143,2.37,2.413793103448276,3.072463768115942,2.58,2.175, Yi-6B,2.775,2.76,2.557142857142857,2.3,2.0517241379310347,3.0434782608695654,2.74,1.4125, Llama-2-7b-hf,2.4625,2.87,2.257142857142857,2.05,1.793103448275862,3.1594202898550723,2.4,1.2625, CodeLlama-7b-hf,1.75,2.05,1.4714285714285715,1.59,1.5344827586206897,2.260869565217391,1.79,1.375, Meta-Llama-3-8B,2.975,2.81,2.3142857142857145,2.27,2.3620689655172415,2.9130434782608696,2.64,1.65, llemma_7b,2.2375,2.46,1.8285714285714285,1.97,1.896551724137931,2.5217391304347827,2.03,1.6125, OLMo-7B,2.125,2.19,1.7428571428571429,1.76,1.8275862068965518,2.6666666666666665,2.02,1.15, gemma-7b-it,3.15,3.34,2.8142857142857145,2.91,2.8275862068965516,3.652173913043478,3.17,2.2,1.6571428571428573 gemma-1.1-7b-it,3.4875,3.56,3.3142857142857145,3.12,3.0517241379310347,4.072463768115942,3.44,2.675,2.0285714285714285 Mistral-7B-Instruct-v0.2,3.6875,3.74,3.6,3.01,3.103448275862069,3.9565217391304346,3.49,3.0125,2.6 Qwen1.5-7B-Chat,3.4,3.74,3.4,3.04,3.0,3.753623188405797,3.71,2.975,2.0428571428571427 Yi-6B-Chat,3.0,3.45,3.1285714285714286,2.49,2.603448275862069,3.5072463768115942,3.56,1.8875,1.5285714285714285 Llama-2-7b-chat-hf,3.4375,3.62,3.3714285714285714,2.64,2.7413793103448274,4.260869565217392,3.58,2.175,2.085714285714286 CodeLlama-7b-Instruct-hf,3.1375,3.18,3.0285714285714285,2.58,2.586206896551724,3.8260869565217392,3.19,2.2125,1.7 Meta-Llama-3-8B-Instruct,3.85,3.75,3.8142857142857145,3.3,3.3448275862068964,3.927536231884058,3.71,3.3625,3.0428571428571427 OLMo-7B-SFT,2.8625,3.13,2.8857142857142857,2.33,2.2586206896551726,3.5072463768115942,2.95,1.725,1.2285714285714286 OLMo-7B-Instruct,2.95,3.44,2.9714285714285715,2.33,2.413793103448276,3.072463768115942,3.19,1.9875,1.4 tulu-2-7b,2.85,3.21,3.1,2.56,2.5172413793103448,3.681159420289855,3.12,2.0,1.7285714285714286 tulu-2-dpo-7b,3.25,3.67,3.242857142857143,2.68,2.706896551724138,3.7681159420289854,3.51,2.325,1.9857142857142858 codetulu-2-7b,2.8,3.18,3.0,2.49,2.7241379310344827,3.347826086956522,3.12,2.525,1.8285714285714285 Orca-2-7b,2.2875,2.26,1.3142857142857143,1.72,1.8103448275862069,2.6231884057971016,2.25,1.3375,1.8428571428571427 openchat-3.5-0106,3.525,3.76,3.5142857142857142,3.26,3.310344827586207,3.8405797101449277,3.61,2.8875,2.3142857142857145 OpenHermes-2-Mistral-7B,3.25,3.55,3.642857142857143,2.89,2.8448275862068964,3.4927536231884058,3.32,2.6375,1.9714285714285715 OpenHermes-2.5-Mistral-7B,3.575,3.53,3.557142857142857,3.07,3.1724137931034484,3.3043478260869565,3.42,2.875,2.242857142857143 Nous-Hermes-2-Mistral-7B-DPO,3.4375,3.58,3.6285714285714286,3.05,3.1724137931034484,3.318840579710145,3.46,2.925,2.2142857142857144 Starling-LM-7B-alpha,3.7125,3.74,3.5,3.2,2.9482758620689653,3.9420289855072466,3.53,2.8375,2.1285714285714286 Starling-LM-7B-beta,3.775,3.86,3.8,3.44,3.5344827586206895,3.9855072463768115,3.91,3.325,2.4285714285714284 mistral-orpo-alpha,3.3875,3.56,3.442857142857143,2.86,3.103448275862069,4.028985507246377,3.45,2.825,2.1142857142857143 mistral-orpo-beta,3.4625,3.66,3.4285714285714284,2.97,2.9310344827586206,3.898550724637681,3.54,2.8125,2.1285714285714286 zephyr-7b-beta,3.375,3.56,3.5,3.0,2.896551724137931,3.5217391304347827,3.5,3.05,1.957142857142857 Qwen1.5-14B,3.3875,3.3,2.914285714285714,2.72,2.8620689655172415,2.6231884057971016,3.06,2.55, Llama-2-13b-hf,2.7625,3.01,2.6,2.15,2.1379310344827585,3.217391304347826,2.65,1.5125, CodeLlama-13b-hf,2.1,2.06,1.7571428571428571,1.71,1.6206896551724137,2.2753623188405796,1.89,1.5875, SOLAR-10.7B-v1.0,3.0875,3.37,3.1142857142857143,2.75,2.7586206896551726,3.5652173913043477,3.25,2.225, Qwen1.5-14B-Chat,3.5875,3.77,3.6142857142857143,3.26,3.1206896551724137,3.8840579710144927,3.5,3.0625,2.4857142857142858 SOLAR-10.7B-Instruct-v1.0,3.7,3.8,3.585714285714286,3.21,3.0344827586206895,3.8260869565217392,3.7,3.4875,2.585714285714286 aya-101,1.25,1.4,1.3571428571428572,1.34,1.3620689655172413,1.6666666666666667,1.4,1.15,1.1571428571428573 Llama-2-13b-chat-hf,3.5875,3.7,3.342857142857143,2.71,2.8620689655172415,4.318840579710145,3.66,2.5125,2.342857142857143 CodeLlama-13b-Instruct-hf,3.0375,3.2,3.157142857142857,2.59,2.4827586206896552,3.971014492753623,3.21,2.3125,2.157142857142857 tulu-2-13b,3.0125,3.31,3.2714285714285714,2.68,2.706896551724138,3.8405797101449277,3.2,2.325,2.057142857142857 tulu-2-dpo-13b,3.4125,3.58,3.4571428571428573,2.71,3.0344827586206895,3.8840579710144927,3.55,2.775,2.2285714285714286 codetulu-2-13b,3.0875,3.37,3.057142857142857,2.62,2.793103448275862,3.420289855072464,3.22,2.9875,1.8 Orca-2-13b,2.8875,2.47,1.6285714285714286,2.13,2.0172413793103448,2.8260869565217392,2.8,2.05,1.9714285714285715 Yi-34B,3.4875,3.37,3.1857142857142855,3.05,2.8793103448275863,3.681159420289855,3.21,2.1625, llemma_34b,2.8375,2.8,2.5,2.53,2.2758620689655173,2.8840579710144927,2.61,1.775, Qwen1.5-32B,3.125,3.52,3.142857142857143,2.99,2.810344827586207,3.536231884057971,3.07,2.6375, CodeLlama-34b-hf,2.675,2.41,2.1285714285714286,1.98,2.0689655172413794,2.5942028985507246,2.45,1.8, Mixtral-8x7B-v0.1,3.55,3.45,3.1857142857142855,3.14,2.7586206896551726,3.8115942028985508,3.33,2.5375, Yi-34B-Chat,3.4625,3.74,3.7142857142857144,3.27,3.413793103448276,4.086956521739131,3.81,2.8125,2.0142857142857142 Nous-Hermes-2-Yi-34B,3.2,3.63,3.557142857142857,3.24,3.206896551724138,3.608695652173913,3.55,2.85,1.9 CodeLlama-34b-Instruct-hf,3.35,3.39,3.2857142857142856,2.85,2.7241379310344827,4.101449275362318,3.37,2.5,2.1857142857142855 codetulu-2-34b,3.3875,3.4,3.414285714285714,3.01,3.1379310344827585,3.7246376811594204,3.43,3.075,2.0142857142857142 Qwen1.5-32B-Chat,3.65,3.85,3.642857142857143,3.55,3.1206896551724137,4.246376811594203,3.8,3.4875,2.6714285714285713 Mixtral-8x7B-Instruct-v0.1,3.65,3.89,3.5714285714285716,3.45,3.1379310344827585,4.0144927536231885,3.78,3.2,2.742857142857143 Nous-Hermes-2-Mixtral-8x7B-SFT,3.6875,3.69,3.6285714285714286,3.16,3.103448275862069,3.652173913043478,3.59,3.225,2.414285714285714 Nous-Hermes-2-Mixtral-8x7B-DPO,3.6625,3.84,3.6714285714285713,3.24,3.1551724137931036,3.782608695652174,3.71,3.3375,2.5285714285714285 c4ai-command-r-v01,3.7125,3.72,3.642857142857143,3.14,3.189655172413793,4.0144927536231885,3.88,2.95,1.957142857142857 Llama-2-70b-hf,3.2875,3.49,3.1,2.78,2.7586206896551726,3.8550724637681157,3.17,2.45, CodeLlama-70b-hf,2.75,2.42,2.3285714285714287,2.32,1.9655172413793103,2.6956521739130435,2.23,2.025, Mixtral-8x22B-v0.1-AWQ,3.525,3.59,3.5,3.44,3.206896551724138,3.9420289855072466,3.37,2.7625, Meta-Llama-3-70B,3.25,3.22,2.7857142857142856,2.76,2.689655172413793,3.260869565217391,2.92,2.3125, Qwen1.5-72B,3.375,3.41,3.1142857142857143,2.97,2.913793103448276,3.898550724637681,3.17,2.7625, Llama-2-70b-chat-hf,3.6125,3.72,3.657142857142857,2.98,3.1551724137931036,4.463768115942029,3.79,2.8875,2.4285714285714284 CodeLlama-70b-Instruct-hf,2.925,2.51,2.3857142857142857,2.62,2.4482758620689653,4.217391304347826,2.56,1.7375,1.7571428571428571 tulu-2-dpo-70b,3.6375,3.8,3.8,3.17,3.1551724137931036,3.8260869565217392,3.7,3.5,2.4 c4ai-command-r-plus-GPTQ,3.925,4.02,3.857142857142857,3.46,3.5172413793103448,3.927536231884058,3.91,3.425,2.8285714285714287 Meta-Llama-3-70B-Instruct,4.175,3.92,3.9714285714285715,3.76,3.7413793103448274,4.028985507246377,3.97,3.625,3.1142857142857143 Mixtral-8x22B-Instruct-v0.1-AWQ,3.8125,3.96,3.7714285714285714,3.6,3.3793103448275863,4.043478260869565,3.84,3.45,2.757142857142857 zephyr-orpo-141b-A35b-v0.1-AWQ,3.2875,3.62,3.6857142857142855,3.25,3.3448275862068964,3.550724637681159,3.45,3.0625,2.5428571428571427 Qwen1.5-72B-Chat,3.7125,3.92,3.7714285714285714,3.53,3.586206896551724,4.101449275362318,3.92,3.425,2.6285714285714286 qwen-110b-chat,4.075,4.03,4.0,3.83,3.7758620689655173,4.130434782608695,3.96,3.325,2.7714285714285714 gpt-3.5-turbo-1106,3.8125,3.75,3.7142857142857144,3.41,3.2413793103448274,4.086956521739131,3.65,3.0,2.585714285714286 gpt-3.5-turbo-0125,3.8,3.86,3.757142857142857,3.43,3.2586206896551726,3.9565217391304346,3.64,2.9875,2.585714285714286 gpt-4-1106-preview,4.0125,4.21,4.0285714285714285,4.01,4.0344827586206895,4.449275362318841,4.09,3.6,3.4285714285714284 gpt-4-0125-preview,4.1125,4.13,3.9285714285714284,4.15,4.0,4.144927536231884,4.15,3.725,3.3285714285714287 gpt-4-turbo-2024-04-09,4.1125,4.09,3.9857142857142858,3.92,3.8620689655172415,4.115942028985507,4.06,3.6875,3.357142857142857 gpt-4o-2024-05-13,4.175,4.14,4.1,3.98,3.789473684210526,4.235294117647059,4.06,3.7875,3.414285714285714 mistral-medium,3.925,3.91,3.842857142857143,3.82,3.5517241379310347,4.115942028985507,3.91,3.6875,2.9714285714285715 mistral-large,3.9,3.83,3.757142857142857,3.66,3.6379310344827585,3.9565217391304346,3.94,3.7125,2.8714285714285714 gemini-1.0-pro,3.5625,3.65,3.6285714285714286,3.48,3.0689655172413794,3.8840579710144927,3.74,3.0625,2.9857142857142858 gemini-pro-1.5,3.875,3.88,3.8714285714285714,3.83,3.5,4.144927536231884,4.01,3.2875,3.1 gemini-flash-1.5,4.05,3.81,3.742857142857143,3.81,3.310344827586207,4.144927536231884,3.97,3.45,2.7285714285714286 claude-3-haiku-20240307,4.0,3.94,3.9571428571428573,3.58,3.5689655172413794,4.27536231884058,3.93,3.5375,2.8714285714285714 claude-3-sonnet-20240229,3.8625,3.83,3.942857142857143,3.84,3.689655172413793,4.2898550724637685,3.86,3.5,3.0428571428571427 claude-3-opus-20240229,4.075,3.88,4.1571428571428575,3.8,3.7413793103448274,4.434782608695652,4.05,3.425,3.357142857142857