AMA

Free, ad-free, open-data

The interactive AI decision engine

Compare 30+ language and image models across real-world scenarios. Drag the weight sliders, set your budget, see what wins.

Pick a scenario

Each scenario weights benchmarks differently. Switch between them — the leaderboard re-ranks live.

Build your own

A pair-programming assistant for IDE / agent loops. Heavy on coding benchmarks, with a real-world agentic component (SWE-bench) and some weight on cost since coding loops burn tokens. A small slice goes to the recovery-rate reliability axis because real-world agent loops live or die on whether the model self-corrects.

#ModelScenario scoreEst. $/monthContextAcc/$Acc·spdStabilityLiveCBSWE-benchAiderArenaFrontierRecovery
1
Google
vision
85.955% cov.
$1202.0M78.087.9
2
Google
vision
85.850% cov.
$301.0M303
3
OpenAI
vision
84.430% cov.
$144400k64.377.6
4
OpenAI
vision
82.730% cov.
$144400k62.977.6
5
DeepSeek
81.675% cov.
$33128k15786.273.7
6
OpenAI
vision
78.065% cov.
$120400k70.127.7
7
OpenAI
vision
77.655% cov.
$120400k69.777.6
8
Anthropic
vision
77.055% cov.
$1,035200k6.42100.0
9
DeepSeek
76.840% cov.
$16128k285
10
Anthropic
vision
76.630% cov.
$207200k29.490.8
11
Anthropic
vision
76.330% cov.
$1,035200k6.36100.0
12
Moonshot (Kimi)
75.770% cov.
$37256k13277.6
13
Anthropic
vision
75.155% cov.
$207200k28.872.5
14
OpenAI
75.1
$66200k73.847.0
15
Alibaba (Qwen)
74.325% cov.
$10131k358
16
Alibaba (Qwen)
74.370% cov.
$10131k35813.2
17
Anthropic
vision
73.355% cov.
$1,035200k6.1158.0
18
OpenAI
vision
72.825% cov.
$120400k64.9
19
Google
vision
71.780% cov.
$1202.0M63.835.7
20
OpenAI
71.170% cov.
$600200k8.6420.5
21
xAI
vision
69.445% cov.
$345256k16.36.40.0
22
Anthropic
vision
68.150% cov.
$69200k72.6
23
OpenAI
vision
67.850% cov.
$24400k270
24
DeepSeek
67.3
$16128k24165.169.5
25
xAI
64.170% cov.
1.0M
26
Zhipu AI (GLM)
63.030% cov.
$28200k124100.0
27
Zhipu AI (GLM)
62.855% cov.
$28200k124100.0
28
Google
vision
61.385% cov.
$301.0M20152.342.7
29
Anthropic
vision
58.7
$207200k22.05.297.5
30
Anthropic
vision
58.665% cov.
$207200k22.065.9
31
Anthropic
vision
58.3
$1,035200k4.861.9100.0
32
Anthropic
vision
56.930% cov.
$207200k21.36.7
33
xAI
56.940% cov.
$2071.0M21.28.4
34
OpenAI
56.3
$66200k52.433.4
35
Anthropic
vision
53.930% cov.
$1,035200k4.491.8
36
OpenAI
vision
52.550% cov.
$5400k818
37
OpenAI
vision
50.565% cov.
1.0M
38
OpenAI
48.725% cov.
$900200k4.0122.2
39
Google
vision
45.925% cov.
$51.0M43132.4
40
Anthropic
vision
43.7
$207200k15.747.2
41
Meta
vision
42.425% cov.
$1010.0M17514.1
42
Google
vision
42.325% cov.
$752.0M32.640.8
43
OpenAI
42.125% cov.
$180128k14.944.6
44
Google
vision
39.150% cov.
$61.0M25225.2
45
Meta
vision
37.840% cov.
$141.0M10213.0
46
xAI
36.225% cov.
$138131k18.036.0
47
Alibaba (Qwen)
35.525% cov.
$30131k31.23.7
48
Meta
34.825% cov.
$29128k30.88.5
49
Meta
31.325% cov.
$116128k8.310.0
50
Anthropic
31.240% cov.
$55200k31.925.6
51
Meta
31.025% cov.
$29128k25.40.6
52
Mistral
30.025% cov.
$102128k13.51.4
53
OpenAI
vision
26.6
$150128k9.7313.4
54
OpenAI
vision
25.670% cov.
$9128k68.21.9
55
OpenAI
vision
13.855% cov.
$510128k1.400.4
56
Mistral
12.025% cov.
$4066k
57
Anthropic
vision
11.150% cov.
$1,035200k0.9213.8
58
OpenAI
image
10.00% cov.
59
OpenAI
image
10.00% cov.
60
Google
image
10.00% cov.
61
Google
image
10.00% cov.
62
Midjourney
image
10.00% cov.
63
Black Forest Labs
image
10.00% cov.
64
Black Forest Labs
image
10.00% cov.
65
Stability AI
image
10.00% cov.
66
Ideogram
image
10.00% cov.
67
Google
vision
7.90% cov.
$1202.0M

Showing 67 of 67 models. Hover any score to see the raw value and contamination risk. Click a model for its full profile.

Cost vs quality

Models on the Pareto frontier (highlighted) give you the best quality at their cost tier.

Pareto frontierBubble size = context window