ModelHumanity's Last ExamPer taskPriceSpeedSupportsContext
57.1high
$1.99$2 / $10841M
47.8high
$1.24$0.75 / $3.751491M
47thinking
$1.30$2 / $121151M
40.8high
$1.60$0.75 / $3.752091M
36.6thinking
—$0.50 / $32111M
Gemma 4 31BGoogleOpen
23.6thinking
—$0.10 / $0.3455256K
22.5thinking
$0.33$1.25 / $101151M
19.9thinking
——1M
Gemma 4 26B A4BGoogleOpen
19.3thinking
—$0.15 / $0.6055256K
18.8high
$0.19$0.30 / $2.504541M
17.2high
$0.06$0.25 / $1.503661M
Gemma 4 12BGoogleOpen
15.7thinking
—$0.10 / $0.30114256K
13.8thinking
——1M
12.1thinking
—$0.30 / $2.502181M
10.8
—$0.05 / $0.15—256K
6.8thinking
—$0.10 / $0.404341M
4.8
——32K
Gemma 4 E4BGoogleOpen
4.8non-thinking
—$0.02 / $0.1042128K
Gemma 4 E2BGoogleOpen
4.8thinking
—$0.08 / $0.40—128K
4.5non-thinking
——32K
4.2
——32K
Gemma 3 270MGoogleOpen
3.6
——32K