ModelGPQA DiamondPer taskPriceSpeedSupportsContext
96.3xhigh
$2.31$10 / $50681.05M
95.3high
$3.49$0.75 / $3.751321M
94.9high
$1.48$2 / $676500K
94.1thinking
$1.44$2 / $121341M
94.1xhigh
$2.15$1.25 / $4.253011M
94.1max
$1.99$4 / $20811.05M
93.7max
$7.63$10 / $50701M
Claude Opus 5Anthropic
93.7high
$3.61$5 / $25541M
Kimi K3Moonshot AIOpen
93.5max
$2.00$3 / $15551M
GPT-5.5OpenAI
93.5xhigh
$2.10$5 / $30921.05M
Qwen3.8 2.4T A95BAlibaba (Qwen)Open
93.5thinking
$2.16$2 / $641262K
93.1high
$0.76$2 / $665500K
MiniMax M3MiniMaxOpen
92.9thinking
$0.51$0.30 / $1.20109524K
92.8max
$0.67$1.32 / $3.96981M
92.8high
$1.60$0.75 / $3.751971M
Qwen3.8 Max (0902)Alibaba (Qwen)
92.8thinking
$5.41$2 / $641984K
92.6max
$8.75$10 / $50671M
92.5max
$1.40$2 / $121371.05M
Qwen3.7 MaxAlibaba (Qwen)
92.3thinking
$0.93$1.48 / $4.432021M
Qwen3.8 FlashAlibaba (Qwen)Open
92.3thinking
$0.37$0.15 / $0.4752262K
92max
$4.08$5 / $25601M
GPT-5.4OpenAI
92xhigh
—$2.50 / $151191.05M
GLM 5.3Z.aiOpen
91.7max
$2.01$1.40 / $4.40941M
91.5xhigh
—$1.75 / $14105400K
91.4max
—$5 / $25521M
91.2max
$0.28$0.15 / $0.50731M
91.1max
$5.09$2 / $10821M
Kimi K2.6Moonshot AIOpen
91.1thinking
$0.71$0.95 / $498256K
91.1max
$0.18$0.20 / $1.201251.05M
Qwen3.8 27BAlibaba (Qwen)Open
90.5xhigh
$1.15$0.42 / $2.5544262K
90.5
—$0.10 / $0.301081M
90.4xhigh
$1.29$1.25 / $4.252921M
GPT-5.2OpenAI
90.3xhigh
—$1.75 / $1481400K
90.1high
$0.21$1.25 / $2.50961M
Qwen3.7 PlusAlibaba (Qwen)
90thinking
$0.22$0.32 / $1.28571M
89.8thinking
—$0.50 / $32121M
89.8xhigh
$1.69$1.25 / $4.252851M
Hy3TencentOpen
89.7thinking
$0.072$0.13 / $0.5384256K
89.6max
—$5 / $25431M
Kimi K2.7 CodeMoonshot AIOpen
89.6thinking
$0.54$0.95 / $483256K
TInkling SmallThinking MachinesOpen
89.5thinking
$0.089$0.45 / $1.20224256K
89.5thinking
—$1 / $295256K
GLM 5.2Z.aiOpen
89.5max
$1.28$1.40 / $4.40941M
Qwen3.5 397B A17BAlibaba (Qwen)Open
89.3thinking
$0.47$0.60 / $3.6058262K
nNex N2 Pronex-agiOpen
89.2
—$0.50 / $2.50—262K
89.1thinking
$0.71$0.30 / $1.20119524K
89xhigh
$0.84$1.25 / $4.251831M
Qwen3.6 Max PreviewAlibaba (Qwen)
88.8thinking
—$1.30 / $7.80—262K
88.4thinking
—$1.25 / $4.25—1M
Qwen3.6 PlusAlibaba (Qwen)
88.2thinking
—$0.50 / $3551M
Kimi K2.5Moonshot AIOpen
87.9thinking
—$0.60 / $381256K
87.5max
$2.49$3 / $15601M
87.5xhigh
$0.37$0.75 / $4.50250400K
MiniMax M2.7MiniMaxOpen
87.4thinking
$0.11$0.30 / $1.2057205K
GPT-5.1OpenAI
87.3high
—$1.25 / $10138400K
IK2 Horizon 375B A23BInstitute of Foundation ModelsOpen
87.3
—$1 / $41,062524K
TInklingThinking MachinesOpen
87.2xhigh
$0.60$0.95 / $4.05204256K
87.1thinking
——128K
87thinking
—$0.43 / $0.87—1M
MMotif 3 (Beta)Motif Technologies
86.9
——262K