The current draw
The sampled questions remain private while models are evaluated under common conditions. Rankings are based on repeated public attempts rather than a single result.
- Questions held
- 1,661
- Hard set (god-mode)
- 421
- Asked this time
- 132
- Not asked (held out)
- 289
- Seed
- 42
- Fingerprint
- 63700c2cd78d6d2cf1308dac
- Public attempts
- 192
claude-fable-5
98.0
grok-4.6
97.6
gpt-5.6-sol
97.5
claude-fable-5.1
96.9
claude-opus-4.7
96.4
gpt-5.6-terra
96.4
gpt-6-astra
96.3
gpt-5.6-luna-pro
96.2
gpt-5.6-sol-pro
96.0
grok-4.5
95.2
claude-sonnet-4.6
95.0
grok-4.7
94.7
grok-4.3
94.7
claude-opus-4.8
94.6
gpt-5.6-terra-pro
94.3
gpt-4.1-mini
94.2
gpt-5.6-luna
94.2
claude-opus-5.5
94.0
gpt-5.5
94.0
claude-sonnet-5.5
93.9
claude-sonnet-5
93.8
claude-sonnet-4
93.6
gpt-6-astra-pro
93.2
o4-mini
92.9
kimi-k3
92.8
qwen3.7-plus
91.6
qwen3.7-max
91.4
gpt-5.4
90.7
gpt-6.1-sol
88.6
gpt-6-luna-pro
88.5
ember-1
88.2
gpt-6-sol
88.2
qwen3-max
87.9
inkling
87.5
gpt-6-sol-pro
87.5
glm-5.3-flash
87.4
gemini-4-argon
87.3
claude-haiku-5.5
86.7
gpt-5.4-nano
86.7
mistral-medium-3.1
86.6
gpt-6.1-sol-pro
86.5
deepseek-v4-pro
85.7
command-a
85.1
claude-opus-5
85.1
grok-4.20
84.9
gpt-6-luna
84.8
gpt-4.1
84.5
mistral-large-4-0
84.3
devstral-2512
83.6
inkling-small
82.1
gpt-5.5-pro
80.9
glm-5.3
80.7
codestral-2508
79.5
llama-4-maverick
79.4
mimo-v2.6-pro
79.1
deepseek-v4-pro-0813
77.7
mistral-medium-3-5
77.5
mimo-v2.5-pro
77.1
deepseek-v4.1-flash
76.9
llama-3.3-70b-instruct
76.2
deepseek-r1-0528
73.6
gemini-3-flash-preview
73.3
mimo-v2.6-flash
73.2
deepseek-v4-flash
72.5
mimo-v2.5
71.6
gemini-2.5-pro
71.4
gemini-3.1-flash-lite
70.6
deepseek-v4-flash-0731
70.2
qwen3.8-2.4t-a95b
70.2
muse-spark-1.3
70.1
gemini-3.7-flash
69.8
gemini-3.8-flash
69.5
kimi-k2
69.5
kimi-k2.5
69.5
gemini-3.5-flash-lite
66.4
gemini-2.5-flash
65.6
gemini-2.5-flash-lite
65.6
gemini-3.5-flash
65.6
qwen3.8-flash
65.2
gemini-3.6-flash
64.5
qwen3.5-flash-02-23
63.8
kimi-k2.6
61.8
muse-wonder
58.3
gemini-3.1-pro-preview
58.0
mimo-v2.6-pro-ultraspeed
56.4
qwen3.6-flash
55.7
qwen3.7-flash
54.6
qwen3.8-max-0902
54.4
muse-glimmer-30b
51.8
command-a-plus
50.8
qwen3.8-max-prime
50.5
glm-5.3-prime
46.4
qwen3.8-27b
45.8
mistral-small-2603
27.1
nova-premier-v1
2.0