Category
Benchmark
Muse Spark Open (xhigh)
Inkling (xhigh)
Nemotron 3 Ultra (thinking)
Beam
Mistral Large 4
Qwen3.8-2.4T-A95B (xhigh)
GLM 5.3 (max)
Kimi K3 (max)
Agentic
Knowledge & Reasoning
Coding
Multimodal (with tools)
61.8
89.2
1401
26.1
58.5
25
76.4
60
50.0
80.2
73.6
31.1
76.0
1079
11.2
47.0
2
55.1
–
14.0
34.2
52.3
27.2
63.1
1016
8.9
40.3
0
53.9
–
Not supported
Not supported
Not supported
–
78.7
–
–
49.7
13
80.1
44
Not supported
Not supported
Not supported
–
–
1424
–
54.2
-5
–
62
–
–
–
52.7
84.5
1613
17.3
54.1
4
82.0
57
Not supported
Not supported
Not supported
61.4
84.2
1653
16.4
59.0
14
83.9
67
Not supported
Not supported
Not supported
54.3
82.3
1537
22.2
59.5
20
85.0
68
41.0
85.7
71.9
Subscribe to our newsletter to keep up with the latest AI updates, releases and more.