"Model Name"	"Link"	"Comment"	"Group"	"Med. Len."	"Med. Resp. Len."	"Parameter Size (B)"	"Type"	"Model Type"	"Think"	"Overall"	"KO"	"EN"	"JA"	"ZH"	"PL"	"DE"	"PT"	"ES"	"FR"	"IT"	"RU"	"VI"
"GPT-5 (Reasoning: medium)"	"https://platform.openai.com/docs/models/gpt-5"	"Reasoning: medium"	"GPT"	""	""	""	"Proprietary"	"Think"	"On"	"70.73"	"64.72"	"65.83"	"71.69"	"67.68"	"72.78"	"71.27"	"73.74"	"75.68"	"72.83"	"77.05"	"70.79"	"75.61"
"o3-pro (Reasoning: medium)"	"https://platform.openai.com/docs/models/o3-pro"	"Reasoning: medium"	"GPT"	""	""	""	"Proprietary"	"Think"	"On"	"66.47"	"63.61"	"63.61"	"69.28"	"65.24"	"63.89"	"64.09"	"68.16"	"69.19"	"70.11"	"72.13"	"62.36"	"71.95"
"GPT-5.2 (Reasoning: medium)"	"https://platform.openai.com/docs/models/gpt-5.2"	"Reasoning: medium"	"GPT"	"347.0"	"264.0"	""	"Proprietary"	"Hybrid"	"On"	"66.18"	"61.67"	"61.39"	"69.28"	"64.63"	"68.89"	"66.3"	"70.95"	"63.24"	"68.48"	"70.49"	"70.22"	"68.29"
"GPT-5.1 (Reasoning: medium, verbosity: medium)"	"https://platform.openai.com/docs/models/gpt-5.1"	"Reasoning: medium, verbosity: medium"	"GPT"	""	""	""	"Proprietary"	"Hybrid"	"On"	"64.57"	"57.78"	"62.5"	"65.06"	"62.8"	"65.56"	"60.22"	"65.36"	"68.11"	"74.46"	"70.49"	"67.42"	"63.41"
"Claude 4.5 Opus (think, budget: 16K)"	"https://www.anthropic.com/claude/opus"	"thinking budget: 16K"	"Claude"	""	""	""	"Proprietary"	"Hybrid"	"On"	"63.41"	"59.44"	"60.28"	"66.27"	"64.02"	"66.67"	"65.19"	"63.69"	"62.16"	"63.59"	"64.48"	"65.73"	"67.07"
"Claude 4 Opus (20250514) (think)"	"https://www.anthropic.com/claude/opus"	"version: 20250514"	"Claude"	""	""	""	"Proprietary"	"Hybrid"	"On"	"63.29"	"57.5"	"62.5"	"64.46"	"62.8"	"59.44"	"65.19"	"65.92"	"60.54"	"65.22"	"65.57"	"65.17"	"72.56"
"Claude 4.1 Opus (20250805) (think)"	"https://www.anthropic.com/claude/opus"	"version: 20250805"	"Claude"	""	""	""	"Proprietary"	"Hybrid"	"On"	"63.24"	"58.33"	"61.39"	"60.84"	"64.02"	"61.67"	"66.85"	"68.16"	"61.08"	"65.76"	"66.67"	"65.73"	"65.24"
"GPT-5 mini (Reasoning: medium)"	"https://platform.openai.com/docs/models/gpt-5-mini"	"Reasoning: medium"	"GPT"	""	""	""	"Proprietary"	"Think"	"On"	"62.56"	"57.5"	"56.39"	"62.65"	"62.2"	"63.89"	"60.22"	"66.48"	"67.03"	"70.11"	"67.76"	"66.29"	"60.98"
"Gemini 3 Pro Preview (Thinking Level: High)"	"https://deepmind.google/models/gemini/pro/"	"Thinking Level: High"	"Gemini"	"1930.5"	"378.0"	""	"Proprietary"	"Think"	"On"	"62.48"	"59.44"	"60.56"	"60.24"	"62.2"	"61.67"	"65.19"	"63.13"	"64.32"	"65.76"	"65.57"	"64.04"	"62.2"
"Claude 4 Sonnet (20250514) (think)"	"https://www.anthropic.com/claude/sonnet"	"version: 20250514"	"Claude"	""	""	""	"Proprietary"	"Hybrid"	"On"	"61.8"	"54.17"	"59.17"	"63.86"	"64.63"	"59.44"	"61.33"	"64.8"	"62.16"	"65.22"	"67.21"	"66.29"	"64.02"
"o3"	"https://platform.openai.com/docs/models/o3"	""	"GPT"	""	""	""	"Proprietary"	"Think"	"On"	"60.91"	"57.5"	"59.17"	"61.45"	"58.54"	"61.11"	"64.09"	"60.89"	"62.16"	"63.59"	"65.03"	"54.49"	"68.29"
"Gemini 2.5 Pro"	"https://deepmind.google/models/gemini/pro/"	""	"Gemini"	""	""	""	"Proprietary"	"Think"	"On"	"59.34"	"53.61"	"57.78"	"59.04"	"57.93"	"57.22"	"56.91"	"60.89"	"63.24"	"67.93"	"62.3"	"61.24"	"60.98"
"Gemini 3 Flash Preview (Thinking Level: High)"	"https://deepmind.google/models/gemini/flash/"	"Thinking Level: High"	"Gemini"	"1296.5"	"424.5"	""	"Proprietary"	"Think"	"On"	"59.26"	"53.89"	"57.22"	"61.45"	"57.32"	"56.67"	"61.33"	"57.54"	"58.92"	"64.67"	"67.76"	"60.11"	"61.59"
"GLM-4.7 FP8"	"https://huggingface.co/zai-org/GLM-4.7-FP8"	"temperature: 1.0
top-p: 0.95"	"GLM"	"2252.5"	"328.0"	"358.0"	"Open"	"Hybrid"	"On"	"59.22"	"54.17"	"55.28"	"63.86"	"63.41"	"55.0"	"58.56"	"62.01"	"61.08"	"63.59"	"61.75"	"66.29"	"54.88"
"DeepSeek V3.2 Speciale"	"https://huggingface.co/deepseek-ai/DeepSeek-V3.2-Speciale"	"temperature: 1.0
top-p: 0.95"	"DeepSeek"	"3226.5"	"249.5"	"671.0"	"Open"	"Think"	"On"	"59.14"	"50.83"	"58.06"	"63.25"	"57.93"	"58.89"	"58.56"	"58.66"	"60.0"	"65.22"	"66.12"	"59.55"	"62.2"
"Grok-4"	"https://x.ai/news/grok-4"	"temperature: 0.6
top-p: 0.95"	"Grok"	""	""	""	"Proprietary"	"Think"	"On"	"58.74"	"57.78"	"56.67"	"62.65"	"60.37"	"58.33"	"60.22"	"59.78"	"56.22"	"62.5"	"60.66"	"52.25"	"60.98"
"Gemini 2.5 Flash"	"https://deepmind.google/models/gemini/flash/"	""	"Gemini"	""	""	""	"Proprietary"	"Hybrid"	"On"	"58.62"	"51.11"	"56.39"	"62.05"	"56.71"	"62.78"	"60.77"	"61.45"	"60.0"	"63.04"	"57.92"	"64.04"	"56.71"
"o4-mini"	"https://platform.openai.com/docs/models/o4-mini"	""	"GPT"	""	""	""	"Proprietary"	"Think"	"On"	"57.57"	"54.17"	"55.0"	"62.05"	"59.76"	"52.78"	"58.56"	"63.69"	"55.68"	"57.61"	"60.66"	"56.74"	"60.98"
"Kimi K2 Thinking"	"https://huggingface.co/moonshotai/Kimi-K2-Thinking"	"temperature:1.0
top-p: 0.95"	"moonshot"	"1692.0"	"330.0"	"1000.0"	"Open"	"Think"	"On"	"56.84"	"50.0"	"57.5"	"60.84"	"62.2"	"53.33"	"54.14"	"61.45"	"53.51"	"59.24"	"59.56"	"56.18"	"61.59"
"Qwen3 235B A22B Thinking 2507"	"https://huggingface.co/Qwen/Qwen3-235B-A22B-Thinking-2507"	"temperature: 0.6
top-p: 0.95"	"Qwen"	"2404.5"	"423.0"	"235.0"	"Open"	"Think"	"On"	"55.48"	"49.17"	"53.33"	"56.02"	"58.54"	"50.56"	"62.43"	"60.89"	"52.97"	"56.52"	"60.11"	"53.93"	"60.37"
"GPT-5 nano (Reasoning: medium)"	"https://platform.openai.com/docs/models/gpt-5-nano"	"Reasoning: medium"	"GPT"	""	""	""	"Proprietary"	"Think"	"On"	"55.39"	"51.94"	"53.89"	"57.23"	"53.66"	"55.56"	"58.01"	"59.78"	"54.59"	"56.52"	"59.02"	"57.3"	"51.83"
"GLM-4.5 FP8"	"https://huggingface.co/zai-org/GLM-4.5-FP8"	"temperature: 0.6
top-p: 0.95"	"GLM"	"1442.0"	"604.0"	"358.0"	"Open"	"Hybrid"	"On"	"54.03"	"46.94"	"54.17"	"60.84"	"58.54"	"48.89"	"55.8"	"54.75"	"48.11"	"57.61"	"57.92"	"57.87"	"54.88"
"GLM-4.6 FP8"	"https://huggingface.co/zai-org/GLM-4.6-FP8"	"temperature: 1.0
top-p: 0.95"	"GLM"	"2645.5"	"522.0"	"358.0"	"Open"	"Hybrid"	"On"	"53.3"	"49.17"	"54.17"	"54.22"	"56.71"	"52.22"	"53.04"	"49.16"	"56.76"	"56.52"	"56.28"	"53.93"	"50.61"
"Gemini 2.5 Flash-lite Preview (09-2025)"	"https://deepmind.google/models/gemini/"	"version: 09-2025"	"Gemini"	""	""	""	"Proprietary"	"Think"	"On"	"53.06"	"47.78"	"51.11"	"51.2"	"53.66"	"51.67"	"54.7"	"59.22"	"51.89"	"57.07"	"55.74"	"57.87"	"51.83"
"Qwen3 235B A22B Instruct 2507"	"https://huggingface.co/Qwen/Qwen3-235B-A22B-Instruct-2507"	"temperature: 0.7
top-p: 0.8"	"Qwen"	"433.0"	"433.0"	"235.0"	"Open"	"Instruct"	"Off"	"52.94"	"46.67"	"55.28"	"53.61"	"59.15"	"46.11"	"51.38"	"55.87"	"54.59"	"53.26"	"56.28"	"54.49"	"53.05"
"DeepSeek V3.2"	"https://huggingface.co/deepseek-ai/DeepSeek-V3.2"	"temperature: 1.0
top-p: 0.95"	"DeepSeek"	"762.5"	"312.0"	"671.0"	"Open"	"Think"	"On"	"52.17"	"47.5"	"49.44"	"53.61"	"50.61"	"50.56"	"54.14"	"59.22"	"52.43"	"57.07"	"56.28"	"44.94"	"57.93"
"DeepSeek V3.1 (think)"	"https://huggingface.co/deepseek-ai/DeepSeek-V3.1"	"temperature: 0.6
top-p: 0.95"	"DeepSeek"	"710.5"	"356.0"	"671.0"	"Open"	"Hybrid"	"On"	"51.45"	"44.44"	"48.33"	"56.63"	"48.78"	"48.89"	"55.25"	"53.07"	"52.97"	"56.52"	"57.92"	"50.56"	"54.27"
"DeepSeek V3.1 Terminus (think)"	"https://huggingface.co/deepseek-ai/DeepSeek-V3.1-Terminus"	"temperature: 0.6
top-p: 0.95"	"DeepSeek"	"831.5"	"377.0"	"671.0"	"Open"	"Hybrid"	"On"	"51.37"	"46.94"	"50.83"	"51.81"	"53.66"	"50.0"	"53.59"	"51.96"	"55.14"	"53.8"	"54.64"	"48.31"	"50.61"
"Qwen3 30B A3B Thinking 2507"	"https://huggingface.co/Qwen/Qwen3-30B-A3B-Thinking-2507"	"temperature: 0.6
top-p: 0.95"	"Qwen"	"2830.0"	"351.0"	"30.0"	"Open"	"Think"	"On"	"50.44"	"44.17"	"49.17"	"50.0"	"57.32"	"42.22"	"49.72"	"53.07"	"50.27"	"54.89"	"56.83"	"47.75"	"58.54"
"MiMo V2 Flash"	"https://huggingface.co/XiaomiMiMo/MiMo-V2-Flash"	"temperature: 0.8
top-p: 0.95"	"XiaomiMiMo"	"1477.5"	"373.0"	"309.0"	"Open"	"Think"	"On"	"50.32"	"42.22"	"53.06"	"49.4"	"54.27"	"47.78"	"51.93"	"53.63"	"52.97"	"54.89"	"54.64"	"42.13"	"52.44"
"gpt-oss-120B (Reasoning: medium)"	"https://huggingface.co/openai/gpt-oss-120b"	"Reasoning: medium
temperature: 1.0
top-p: 1.0"	"GPT"	"759.5"	"370.5"	"117.0"	"Open"	"Think"	"On"	"49.11"	"46.67"	"51.39"	"51.81"	"47.56"	"45.0"	"51.38"	"54.75"	"50.27"	"51.63"	"47.54"	"46.07"	"45.12"
"DeepSeek R1 (0528) (top_p: 0.95, temp:0.6)"	"https://huggingface.co/deepseek-ai/DeepSeek-R1-0528"	"version: 0528
temperature: 0.6
top-p: 0.95"	"DeepSeek"	"1177.5"	"554.0"	"671.0"	"Open"	"Think"	"On"	"48.79"	"42.22"	"49.44"	"50.0"	"53.05"	"47.22"	"48.62"	"50.28"	"48.11"	"51.63"	"54.1"	"44.38"	"53.05"
"Gauss2.3 Hybrid"	""	""	"Gauss"	"546.0"	"308.0"	""	"Proprietary"	"Hybrid"	"On"	"46.58"	"39.72"	"45.56"	"48.8"	"48.17"	"45.0"	"44.2"	"53.63"	"45.41"	"52.17"	"51.91"	"44.94"	"47.56"
"Mistral Large 3 675B Instruct 2512"	"https://huggingface.co/mistralai/Mistral-Large-3-675B-Instruct-2512"	"temperature: 0.15"	"mistralai"	"448.0"	"448.0"	"675.0"	"Open"	"Instruct"	"Off"	"45.21"	"41.39"	"44.17"	"50.6"	"46.34"	"46.11"	"43.65"	"45.81"	"44.32"	"49.46"	"49.18"	"42.13"	"44.51"
"DeepSeek V3 (0324) (top_p: 0.95, temp:1.3)"	"https://huggingface.co/deepseek-ai/DeepSeek-V3-0324"	"version: 0324
temperature: 1.3
top-p: 0.95"	"DeepSeek"	"408.0"	"408.0"	"671.0"	"Open"	"Instruct"	"Off"	"45.09"	"37.5"	"43.61"	"46.99"	"51.22"	"45.56"	"44.75"	"44.69"	"44.32"	"48.91"	"49.18"	"44.94"	"49.39"
"Qwen3 32B (think)"	"https://huggingface.co/Qwen/Qwen3-32B"	"temperature: 0.6
top-p: 0.95"	"Qwen"	"1113.0"	"390.0"	"32.8"	"Open"	"Hybrid"	"On"	"44.44"	"38.89"	"41.67"	"48.8"	"50.0"	"38.33"	"46.41"	"44.69"	"44.86"	"44.57"	"50.82"	"46.07"	"47.56"
"Qwen3 30B A3B Instruct 2507"	"https://huggingface.co/Qwen/Qwen3-30B-A3B-Instruct-2507"	"temperature: 0.7
top-p: 0.8"	"Qwen"	"441.5"	"441.5"	"30.0"	"Open"	"Instruct"	"Off"	"42.79"	"34.44"	"43.89"	"40.96"	"48.78"	"38.89"	"41.99"	"46.93"	"44.32"	"42.93"	"48.09"	"43.26"	"46.95"
"MiniMax-M2 (230B A10B)"	"https://huggingface.co/MiniMaxAI/MiniMax-M2"	"temperature:1.0
top-p: 0.95"	"MiniMaxAI"	"1142.0"	"325.0"	"230.0"	"Open"	"Think"	"On"	"42.43"	"31.94"	"46.11"	"37.35"	"45.73"	"38.33"	"45.3"	"45.25"	"48.65"	"41.3"	"46.45"	"42.7"	"46.95"
"A.X 4.0"	"https://huggingface.co/skt/A.X-4.0"	""	"SKT"	"412.5"	"412.5"	"71.9"	"Open"	"Instruct"	"Off"	"41.59"	"38.89"	"41.11"	"43.98"	"49.39"	"36.11"	"45.86"	"43.58"	"44.32"	"39.67"	"43.17"	"39.89"	"36.59"
"gpt-oss-20B (Reasoning: medium)"	"https://huggingface.co/openai/gpt-oss-20b"	"Reasoning: medium
temperature: 1.0
top-p: 1.0"	"GPT"	"953.5"	"326.0"	"21.0"	"Open"	"Think"	"On"	"41.18"	"36.67"	"42.78"	"45.78"	"45.73"	"37.78"	"35.91"	"41.9"	"39.46"	"51.09"	"40.44"	"38.76"	"41.46"
"Gemma 3 27B it"	"https://huggingface.co/google/gemma-3-27b-it"	"temperature: 1.0
top-p: 0.95"	"Gemma"	"380.0"	"380.0"	"27.0"	"Open"	"Instruct"	"Off"	"40.86"	"34.44"	"35.0"	"37.35"	"43.9"	"42.22"	"43.65"	"47.49"	"41.08"	"44.02"	"53.55"	"39.33"	"40.24"
"Tongyi DeepResearch 30B A3B"	"https://huggingface.co/Alibaba-NLP/Tongyi-DeepResearch-30B-A3B"	"temperature: 0.6
top-p: 0.95"	"Alibaba"	"1147.0"	"408.0"	"30.0"	"Open"	"Think"	"On"	"40.1"	"36.11"	"40.83"	"43.37"	"44.51"	"32.78"	"37.02"	"44.69"	"38.92"	"43.48"	"46.45"	"37.08"	"39.63"
"Mistral Small 3.2 24B Instruct 2506"	"https://huggingface.co/mistralai/Mistral-Small-3.2-24B-Instruct-2506"	"temperature: 0.15
top-p: 0.95"	"mistralai"	"369.0"	"369.0"	"24.0"	"Open"	"Instruct"	"Off"	"39.09"	"31.39"	"40.0"	"36.75"	"42.07"	"34.44"	"44.2"	"41.9"	"42.16"	"45.65"	"40.98"	"37.64"	"38.41"
"K2-Think"	"https://huggingface.co/LLM360/K2-Think"	"temperature: 1.0
top-p: 0.95"	"LLM360"	"1835.0"	"486.0"	"32.8"	"Open"	"Think"	"On"	"35.06"	"29.17"	"36.11"	"30.12"	"44.51"	"26.67"	"33.15"	"38.55"	"37.84"	"41.85"	"37.7"	"33.71"	"36.59"
"Kanana 2 30B A3B Thinking"	"https://huggingface.co/kakaocorp/kanana-2-30b-a3b-thinking"	"temperature: 0.6
top-p: 0.95
top-k: 20"	"Kakao"	"4263.0"	"854.5"	"31.0"	"Open"	"Think"	"On"	"34.5"	"25.28"	"43.06"	"38.55"	"40.24"	"25.0"	"34.25"	"37.99"	"32.43"	"34.24"	"37.7"	"28.65"	"38.41"
"KAT Dev 72B Exp"	"https://huggingface.co/Kwaipilot/KAT-Dev-72B-Exp"	"temperature:0.6
top-p: 0.95"	"KAT"	"397.0"	"397.0"	"72.0"	"Open"	"Instruct"	"Off"	"33.94"	"25.0"	"32.22"	"31.93"	"37.2"	"34.44"	"33.15"	"43.02"	"37.84"	"36.96"	"37.7"	"30.34"	"38.41"
"Olmo 3 32B Think"	"https://huggingface.co/allenai/Olmo-3-32B-Think"	"temperature: 1
top-p: 0.95
top-k: 50"	"allenai"	"3360.5"	"473.0"	"32.0"	"Open"	"Think"	"On"	"33.94"	"30.56"	"41.39"	"30.12"	"31.1"	"25.0"	"34.25"	"35.75"	"33.51"	"36.41"	"37.16"	"31.46"	"35.98"
"EXAONE 4.0 32B (think)"	"https://huggingface.co/LGAI-EXAONE/EXAONE-4.0-32B"	"temperature: 0.6
top-p: 0.95"	"Exaone"	"1274.5"	"503.0"	"32.0"	"Open"	"Hybrid"	"On"	"33.82"	"33.61"	"38.33"	"28.92"	"35.98"	"26.11"	"35.91"	"34.08"	"38.92"	"35.33"	"33.88"	"28.09"	"31.71"
"Apriel 1.5 15B Thinker"	"https://huggingface.co/ServiceNow-AI/Apriel-1.5-15b-Thinker"	"temperature: 0.6
top-p: 0.95"	"Apriel"	"2238.0"	"375.0"	"15.0"	"Open"	"Think"	"On"	"31.92"	"23.61"	"39.72"	"30.72"	"38.41"	"24.44"	"40.88"	"37.99"	"32.43"	"32.61"	"22.95"	"28.65"	"31.71"
"HyperCLOVAX SEED Think 14B (think)"	"https://huggingface.co/naver-hyperclovax/HyperCLOVAX-SEED-Think-14B"	"temperature: 0.5
top-p: 0.6"	"HCX"	"1444.0"	"382.5"	"14.7"	"Open"	"Hybrid"	"On"	"31.84"	"32.22"	"37.22"	"31.93"	"38.41"	"27.78"	"32.6"	"30.17"	"29.19"	"32.07"	"33.33"	"25.28"	"26.22"
"Kanana 2 30B A3B Instruct"	"https://huggingface.co/kakaocorp/kanana-2-30b-a3b-instruct"	"temperature: 0"	"Kakao"	"1195.0"	"1195.0"	"31.0"	"Open"	"Instruct"	"Off"	"30.84"	"33.06"	"39.44"	"37.35"	"33.54"	"17.78"	"26.52"	"25.14"	"30.81"	"29.35"	"31.15"	"23.03"	"32.93"
"Dhanishtha-2.0 Preview"	"https://huggingface.co/HelpingAI/Dhanishtha-2.0-preview"	"temperature: 0.7
top-p: 0.9"	"HelpingAI"	"520.0"	"356.0"	"14.8"	"Open"	"Think"	"On"	"25.81"	"23.33"	"27.22"	"30.12"	"32.32"	"20.56"	"20.99"	"26.26"	"25.95"	"25.54"	"30.6"	"23.6"	"25.0"
"ERNIE 4.5 21B A3B Thinking"	"https://huggingface.co/baidu/ERNIE-4.5-21B-A3B-Thinking"	"temperature: 0.6
top-p: 0.95"	"ERNIE"	"1637.0"	"541.0"	"21.0"	"Open"	"Think"	"On"	"25.32"	"17.5"	"31.11"	"18.67"	"39.02"	"23.33"	"24.31"	"24.58"	"26.49"	"24.46"	"30.6"	"19.1"	"27.44"
"Solar Pro Preview (top_p:0.95, temp: 0.7)"	"https://huggingface.co/upstage/solar-pro-preview-instruct"	"temperature: 0.7
top-p: 0.95"	"Solar"	"260.0"	"260.0"	"22.0"	"Open"	"Instruct"	"Off"	"20.73"	"9.72"	"22.22"	"21.08"	"24.39"	"9.44"	"18.23"	"24.02"	"29.73"	"29.89"	"33.33"	"22.47"	"12.8"
"Mi:dm 2.0 Base Instruct"	"https://huggingface.co/K-intelligence/Midm-2.0-Base-Instruct"	"temperature: 0.8
top-p: 0.7"	"KT"	"316.0"	"316.0"	"11.5"	"Open"	"Instruct"	"Off"	"20.25"	"26.39"	"26.39"	"17.47"	"26.83"	"13.33"	"18.78"	"20.67"	"16.22"	"20.65"	"21.31"	"12.92"	"9.15"
"Kanana 1.5 15.7B A3B Instruct"	"https://huggingface.co/kakaocorp/kanana-1.5-15.7b-a3b-instruct"	"temperature: 1.0
top-p: 0.95"	"Kakao"	"414.0"	"414.0"	"15.7"	"Open"	"Instruct"	"Off"	"11.71"	"21.11"	"20.28"	"10.84"	"15.24"	"5.56"	"7.73"	"8.94"	"9.19"	"8.15"	"5.46"	"5.06"	"4.88"