AI RACE— The AI Race
← All rankings

🧠 Reasoning

Reasoning capabilities & academic knowledge — GPQA Diamond, MMLU-Pro, HLE.

100/100 models
#Access
199.4Try now
298.4Try now
3
SpaceXAI
97.5–
497.2Try now
597.2Try now
696.8Try now
796.2Try now
896.1Try now
9
Anthropic
96.0Try now
1095.9Try now
1195.6Try now
1295.5Try now
1395.3Try now
14
SpaceXAI
95.1Try now
1595.0Try now
1694.5Try now
1794.5Try now
1894.2Try now
1994.2Try now
20
MiniMax
94.1Try now
2194.0Try now
22
OpenAI
94.0Try now
2393.3Try now
2493.3Try now
2593.3Try now
2693.3Try now
2793.2Try now
2893.2Try now
2993.1Try now
3093.0Try now
3193.0Try now
32
Thinking Machines
92.0–
3391.9Try now
3491.7Try now
35
91.7–
36
Tencent
91.7–
3791.5Try now
38
OpenAI
91.5Try now
3991.3Try now
4091.3–
4191.1Try now
4291.1–
4391.1Try now
4491.0Try now
45
SpaceXAI
90.8–
46
Nex AGI
90.5–
47
Institute of Foundation Models
90.5–
4890.4Try now
4989.9Try now
50
Motif Technologies
89.7–
51
89.6–
52
89.6–
53
Thinking Machines
89.4–
5489.3Try now
5588.6Try now
5688.6Try now
57
SK Telecom
88.5–
5888.5–
5988.5Try now
6088.3Try now
6188.2Try now
6288.2Try now
63
InclusionAI
88.2–
64
OpenAI
88.1Try now
6588.1Try now
6687.8Try now
6787.8Try now
6887.3Try now
69
87.3–
7087.2Try now
7187.2–
7287.0Try now
7386.8–
7486.8Try now
7586.7Try now
7686.6–
77
InclusionAI
86.6–
7886.5Try now
7986.3Try now
8086.2Try now
8185.9Try now
82
LG AI Research
85.8–
8385.6–
8485.6–
85
Institute of Foundation Models
85.5–
86
Xiaomi
85.5–
8785.2Try now
8885.1Try now
8984.9Try now
9084.8Try now
9184.8Try now
92
OpenAI
84.6Try now
9384.6Try now
9484.5Try now
9584.2Try now
9684.1–
9784.0Try now
9884.0Try now
99
China Mobile
84.0–
10083.9Try now

How scores work — AIM Index

● Capability 85%● Recency 15%
GPQA DiamondARC-AGIHLE

Data sources (1)