ModelMathArena · Apex ShortlistPer taskPriceSpeedSupportsContext
GPT-5.5OpenAI
98xhigh
$2.63$5 / $301131.05M
92unstated
—$2 / $12—1M
Kimi K3Moonshot AIOpen
92unstated
—$3 / $15—1M
90max
$4.08$5 / $25661M
89max
—$0.30 / $1.20—1M
88max
—$1.32 / $3.96—1M
87high
—$5 / $25451M
GPT-5.4OpenAI
81xhigh
—$2.50 / $151271.05M
GPT-5.2OpenAI
79high
—$1.75 / $14—400K
Kimi K2.6Moonshot AIOpen
77unstated
—$0.95 / $4—256K
Step 3.7 FlashStepFunOpen
77unstated
—$0.20 / $1.15—256K
GLM 5.1Z.aiOpen
72unstated
—$1.40 / $4.40—203K
71unstated
—$0.75 / $3.75—1M
70unstated
——128K
69unstated
—$0.50 / $3—1M
GLM 5Z.aiOpen
69unstated
—$1 / $3.20—203K
GLM 5.2Z.aiOpen
68unstated
—$1.40 / $4.40—1M
64xhigh
—$5 / $25—1M
Qwen3.5 397B A17BAlibaba (Qwen)Open
60unstated
—$0.60 / $3.60—262K
Kimi K2.5Moonshot AIOpen
57unstated
—$0.60 / $3—256K
GPT-5.1OpenAI
56high
—$1.25 / $10128400K
Qwen 3.5 27BAlibaba (Qwen)Open
52unstated
—$0.30 / $2.40—262K
Kimi K2 ThinkingMoonshot AIOpen
47unstated
—$0.47 / $2—256K
GPT OSS 120BOpenAIOpen
45unstated
—$0.03 / $0.15—131K
Qwen 3.5 35B A3BAlibaba (Qwen)Open
45unstated
—$0.25 / $2—262K
40unstated
—$0.25 / $2—400K
Qwen3.5 4BAlibaba (Qwen)Open
33unstated
—$0.03 / $0.15—262K
Qwen 3.5 9BAlibaba (Qwen)Open
30unstated
—$0.08 / $0.13—262K
lQED Nanolm-proversOpen
22
—$0.40 / $0.80—262K
Qwen3.5 2BAlibaba (Qwen)Open
8unstated
—$0.02 / $0.10—262K