ModelAutomationBench-AAPer taskPriceSpeedSupportsContext
77.5high
$1.99$2 / $10841M
72.5thinking
$0.22$0.15 / $0.601051M
71.8max
$5.46$2 / $101371M
71.3
——1M
69.5max
$5.98$4 / $20961M
68.5max
$3.26$10 / $50691.05M
68.1max
$4.52$4 / $201611M
67xhigh
$1.75$2 / $665500K
66.6max
$0.26$0.30 / $1.201671M
66.6xhigh
$0.39$2 / $10531.05M
65.6xhigh
$3.74$2 / $694500K
MiMo v2.6 FlashXiaomiOpen
64.1thinking
$0.062$0.14 / $0.28411M
GLM 5.3Z.aiOpen
62.2max
$2.01$1.40 / $4.401861M
Ling 3.1 FlashinclusionAIOpen
61.7
$1.00$0.30 / $0.901451M
GPT-6 SolOpenAI
61.7xhigh
$0.52$2 / $101171.05M
60.9medium
$0.93$0.75 / $3.75—1M
60.4max
$0.28$0.15 / $0.501791M
60.1max
$1.99$4 / $201141.05M
59.9thinking
$1.13$1.36 / $4.18671M
59.6max
$1.40$2 / $121421.05M
59.4max
$7.63$10 / $50691M
MiMo v2.6 ProXiaomiOpen
58.6thinking
$0.13$0.43 / $0.87541M
Kimi K3Moonshot AIOpen
58.3max
$2.00$3 / $151121M
57.9high
$0.88$2 / $663500K
57.9max
$1.60$1.25 / $4.251281M
Qwen3.8 2.4T A95BAlibaba (Qwen)Open
57.2thinking
$2.16$2 / $6127262K
56.7max
$0.67$1.32 / $3.96901M
Claude Opus 5Anthropic
56.6max
$5.86$5 / $25551M
Qwen3.8 Max (0902)Alibaba (Qwen)
56.2thinking
$5.41$2 / $640984K
Qwen3.8 FlashAlibaba (Qwen)Open
55.9thinking
$0.37$0.15 / $0.4755262K
54.1max
$8.75$10 / $50671M
53.2max
$0.068$0.10 / $0.501381.05M
53high
$1.60$0.75 / $3.752091M
Step 5 PreviewStepFunOpen
51thinking
$1.03$1 / $2.70931M
50.2max
$0.18$0.20 / $1.201781.05M
49.7xhigh
$0.84$1.25 / $4.251261M
RBeam 1.0 501B A23BReflectionOpen
48.5thinking
$0.57$0.25 / $13331M
Qwen3.8 27BAlibaba (Qwen)Open
48.2xhigh
$1.01$0.42 / $2.55114262K
GPT-5.5OpenAI
47.3xhigh
$2.63$5 / $301131.05M
46.5thinking
——262K
45.6max
$4.08$5 / $25661M
40.6xhigh
$0.97$1.25 / $4.253551M
38.8xhigh
$1.38$1.25 / $4.252801M
IK2 Horizon 375B A23BInstitute of Foundation ModelsOpen
37.2
—$1 / $41,109524K
36.5max
$5.09$2 / $10801M
36.4
$0.69167400K
36xhigh
$0.12$0.10 / $0.501961M
35.4thinking
$1.30$2 / $121151M
GLM 5.2Z.aiOpen
28.4max
$1.47$1.40 / $4.401891M
AApodex 1.1Apodex
28.2
$0.46$0.30 / $3—256K
26.2xhigh
$0.45$0.75 / $4.50307400K
IK2 Horizon MoVA 36B A4BInstitute of Foundation ModelsOpen
26.1thinking
—$1 / $4—524K
25high
$0.19$0.30 / $2.504541M
Kimi K2.7 CodeMoonshot AIOpen
24.5thinking
$0.54$0.95 / $463256K
MQuasar 438BMultiverse Computing
23.9
$2.02$0.60 / $1.801661M
Qwen3.7 MaxAlibaba (Qwen)
23thinking
$1.15$1.48 / $4.432011M
22.3
$0.37$0.05 / $0.20129524K
MiniMax M3MiniMaxOpen
21.3thinking
$0.51$0.30 / $1.20157524K
IK2 Horizon 7BInstitute of Foundation ModelsOpen
20.9thinking
—$0.30 / $1.20—524K
GLM 5.1Z.aiOpen
20.3thinking
$1.02$1.40 / $4.4072203K