ModelITBench SREPer taskPriceSpeedSupportsContext
56.2max
$1.99$4 / $201141.05M
Step 5 PreviewStepFunOpen
55.6thinking
$1.03$1 / $2.70931M
52.5high
$1.24$0.75 / $3.751491M
51.2max
$0.28$0.15 / $0.501791M
51max
$1.40$2 / $121421.05M
49.5max
$7.63$10 / $50691M
GPT-6 SolOpenAI
49.4max
$1.04$2 / $101281.05M
48.6max
$3.26$10 / $50691.05M
Kimi K3Moonshot AIOpen
47.7max
$2.00$3 / $151121M
46.7max
—$5 / $25631M
GLM 5.3Z.aiOpen
46.1max
$2.01$1.40 / $4.401861M
GPT-5.5OpenAI
45.8xhigh
$2.63$5 / $301131.05M
GLM 5.2Z.aiOpen
42.7max
$1.47$1.40 / $4.401891M
Qwen3.7 MaxAlibaba (Qwen)
42.5thinking
$1.15$1.48 / $4.432011M
42.1xhigh
$3.74$2 / $694500K
40.3max
$0.18$0.20 / $1.201781.05M
GLM 5.1Z.aiOpen
40.3thinking
$1.02$1.40 / $4.4072203K
Qwen3.8 Max (0902)Alibaba (Qwen)
40.3thinking
$5.41$2 / $640984K
39.8max
$2.49$3 / $15621M
38.2max
$5.98$4 / $20961M
MiMo v2.5 ProXiaomiOpen
38.2thinking
$0.054$0.43 / $0.87461M
Gemma 4 31BGoogleOpen
37.3thinking
—$0.10 / $0.3455256K
Qwen 3.5 27BAlibaba (Qwen)Open
35.5thinking
—$0.30 / $2.4062262K
35.2xhigh
$0.45$0.75 / $4.50307400K
Qwen3.5 397B A17BAlibaba (Qwen)Open
34.1thinking
$0.47$0.60 / $3.6055262K
33.2max
$1.60$1.25 / $4.251281M
32.7high
$0.21$1.25 / $2.501881M
Kimi K2.6Moonshot AIOpen
31.2thinking
$0.80$0.95 / $4109256K
30.3thinking
$1.30$2 / $121151M
Step 3.7 FlashStepFunOpen
30.3high
—$0.20 / $1.15123256K
27.3thinking
$0.28$1 / $5105200K
MiniMax M2.7MiniMaxOpen
26.5thinking
—$0.30 / $1.2070205K
24.4xhigh
$0.18$0.20 / $1.25163400K
Gemma 4 26B A4BGoogleOpen
23.6thinking
—$0.15 / $0.6055256K
Qwen 3.5 35B A3BAlibaba (Qwen)Open
21.5thinking
—$0.25 / $281262K
GPT OSS 120BOpenAIOpen
5.6high
$0.092$0.03 / $0.15243131K
1.1
$1.64$0.13 / $1.151481M