ModelGDPval-AA 2.1Per taskPriceSpeedSupportsContext
68.3max
$5.98$4 / $20961M
67.9
——1M
66.9max
$5.46$2 / $101371M
63.5max
$4.52$4 / $201611M
62.7max
$7.63$10 / $50691M
62.3max
$0.26$0.30 / $1.201671M
Claude Opus 5Anthropic
61.1max
$5.86$5 / $25551M
60.6xhigh
$3.74$2 / $694500K
MiMo v2.6 ProXiaomiOpen
59.2thinking
$0.13$0.43 / $0.87541M
59max
$1.60$1.25 / $4.251281M
Qwen3.8 Max (0902)Alibaba (Qwen)
58.2thinking
$5.41$2 / $640984K
GLM 5.3Z.aiOpen
57.5max
$2.01$1.40 / $4.401861M
57.3xhigh
$1.75$2 / $665500K
57.1max
$0.28$0.15 / $0.501791M
Qwen3.8 FlashAlibaba (Qwen)Open
56.4thinking
$0.37$0.15 / $0.4755262K
56.2high
$1.99$2 / $10841M
55.9max
$0.21$0.10 / $0.502341M
Ling 3.1 FlashinclusionAIOpen
55.9
$1.00$0.30 / $0.901451M
55.5max
$8.75$10 / $50671M
55.4max
$1.99$4 / $201141.05M
Qwen3.8 2.4T A95BAlibaba (Qwen)Open
55.4thinking
$2.16$2 / $6127262K
MiMo v2.6 FlashXiaomiOpen
55.4thinking
$0.062$0.14 / $0.28411M
54.7max
$0.72$2 / $10521.05M
Step 5 PreviewStepFunOpen
54.3thinking
$1.03$1 / $2.70931M
53.6max
$3.26$10 / $50691.05M
Kimi K3Moonshot AIOpen
51.6max
$2.00$3 / $151121M
GPT-6 SolOpenAI
50.4max
$1.04$2 / $101281.05M
48.6xhigh
$0.97$1.25 / $4.253551M
48.2max
$5.09$2 / $10801M
48max
$0.18$0.20 / $1.201781.05M
47.9thinking
$0.22$0.15 / $0.601051M
47.7max
$4.08$5 / $25661M
47.6max
$1.40$2 / $121421.05M
47.4max
$0.67$1.32 / $3.96901M
46.8high
$1.24$0.75 / $3.751491M
46.6max
$0.068$0.10 / $0.501381.05M
46.1thinking
$1.13$1.36 / $4.18671M
Qwen3.8 27BAlibaba (Qwen)Open
45.8xhigh
$1.01$0.42 / $2.55114262K
45.1xhigh
$0.84$1.25 / $4.251261M
44.2high
$0.88$2 / $663500K
GLM 5.2Z.aiOpen
43.6max
$1.47$1.40 / $4.401891M
IK2 Horizon 375B A23BInstitute of Foundation ModelsOpen
42.9
—$1 / $41,109524K
42.6max
—$5 / $25631M
GPT-5.5OpenAI
42.5xhigh
$2.63$5 / $301131.05M
41.4
—$0.10 / $0.301171M
41.2thinking
——262K
39.2high
$1.60$0.75 / $3.752091M
GPT-5.4OpenAI
37.3xhigh
—$2.50 / $151271.05M
MiniMax M3MiniMaxOpen
37.1thinking
$0.51$0.30 / $1.20157524K
36.6max
$2.49$3 / $15621M
35.6xhigh
$1.38$1.25 / $4.252801M
IK2 Horizon MoVA 36B A4BInstitute of Foundation ModelsOpen
34.6thinking
—$1 / $4—524K
AApodex 1.1Apodex
34.6
$0.46$0.30 / $3—256K
MQuasar 438BMultiverse Computing
34.1
$2.02$0.60 / $1.801661M
RBeam 1.0 501B A23BReflectionOpen
33.9thinking
$0.57$0.25 / $13331M
33.1
$0.19$0.10 / $0.404131M
Ling 3.0 Flash VLinclusionAIOpen
33.1thinking
—$0.06 / $0.18—262K
Qwen3.7 MaxAlibaba (Qwen)
31.4thinking
$1.15$1.48 / $4.432011M
TInkling SmallThinking MachinesOpen
31.2thinking
$0.089$0.45 / $1.20219256K
MiMo v2.5 ProXiaomiOpen
31thinking
$0.054$0.43 / $0.87461M