UABUnbiased AI BenchAI model rankings with source links.
Every score links back to its source.
Home/Heatmap
Heatmap
Live · updated continuously
Home · heatmap explorer

The AI score matrix,
with sources attached.

Scan many benchmark scores at once, then click any score to see the source, raw value, date, and fair comparison set.
View · score matrix
Sources · 9
Theme · light / dark
Data version

Read this before trusting a headline.

Data version May 13, 2026Model list checked9 providers · 800 tracked modelsPage refreshed May 18, 2026

If this date looks stale, you may be seeing an older build or cached deploy.

Sort models by
Cell shows
Modality
Coverage floor
770 models · 40 benchmarks
AR · ratingAR · ratingAR · ratingAR · ratingAR · ratingAR · ratingAR · ratingAR · ratingAR · ratingAR · ratingAR · ratingLB · %LB · %LB · %LB · %LB · %LB · %AA · indexAA · sAA · %AA · ratingAA · ratingAA · ratingAA · ratingBB · %BB · %BB · %BB · t/sBB · msTERMINAL-BENCH · %SL · %SL · %SL · %SL · %SL · %SL · %SL · %SL · %OC · %MTEB · ndcg
Claude Opus 4.6
Anthropic
100%
98.4%
98.1%
98.4%
100%
100%
95.9%
13.2%
95.8%
72%
100%
40%
60%
86.7%
33.3%
42.9%
30%
81.8%
100%
60%
100%
70%
42.9%
43%
GPT-5.4
OpenAI
96.2%
54.7%
68.3%
54.7%
67.9%
78.9%
79%
0%
20%
62.5%
88%
83.3%
32%
96%
70%
60%
71.4%
100%
100%
50%
20%
84.6%
90%
71.4%
70%
41.7%
Gemini 2.5 Pro
Google
96.8%
7.8%
86.5%
7.8%
28.6%
36.8%
99.2%
99.2%
98.5%
86.4%
100%
100%
68.6%
9.4%
40%
86.7%
100%
90%
27.3%
33.3%
38.5%
10%
100%
60%
37.8%
GPT-5.4 mini
OpenAI
68.1%
67.2%
56.7%
67.2%
14.3%
50.2%
1.9%
20%
45.8%
84%
55.6%
88%
100%
36.7%
60%
71.4%
60%
45.5%
66.7%
46.2%
40%
71.4%
70%
32.2%
Claude Opus 4.7
Anthropic
99.1%
100%
100%
100%
92.9%
94.7%
99.3%
14.2%
83.3%
44%
50%
60%
88%
53.3%
33.3%
42.9%
80%
42.9%
31.9%
GPT-5.5
OpenAI
97.5%
76.6%
96.2%
76.6%
96.4%
84.2%
89.3%
3.3%
33.3%
96%
66.7%
68%
80%
100%
100%
29.1%
Claude Sonnet 4.6
Anthropic
95.9%
95.3%
93.3%
95.3%
85.7%
89.5%
92.3%
2.4%
91.7%
96%
83.3%
52%
68%
13.3%
21.4%
0%
21.4%
27.4%
GPT-5.4 nano
OpenAI
53.3%
62.5%
45.2%
62.5%
14.3%
29.2%
0.9%
20%
41.7%
56%
84%
76%
3.3%
60%
71.4%
60%
45.5%
66.7%
46.2%
40%
71.4%
70%
27%
GPT-5
OpenAI
71%
62.5%
68.3%
62.5%
14.3%
52.4%
3.8%
20%
70%
60%
71.4%
60%
45.5%
66.7%
46.2%
40%
71.4%
70%
23.9%
kimi-k2.6
Unknown
96.2%
93.8%
92.3%
93.8%
68.4%
92.3%
34%
100%
52%
38.9%
48%
36%
21.1%
Grok 4.3
xAI
84.2%
59.4%
86.5%
59.4%
60.7%
70.8%
18.4%
79.2%
76%
77.8%
96%
44%
20.3%
Grok 4.20
xAI
93.7%
65.6%
85.6%
65.6%
64.3%
21.1%
67.5%
5.2%
87.5%
48%
61.1%
92%
52%
0%
20.2%
Claude Haiku 4.5
Anthropic
71.3%
29.7%
36.5%
29.7%
26.3%
78.1%
71.5%
66.9%
50.4%
74%
81.1%
70.8%
87.3%
30%
20.1%
o1
OpenAI
65.6%
64.4%
88.3%
98.5%
97.7%
88.8%
89%
89%
70.8%
6.1%
19%
Claude Sonnet 3.7
Anthropic
49.2%
55.8%
96.9%
97.7%
96.2%
88%
92.1%
98.4%
70.8%
18.6%
qwq-32b
Unknown
54.6%
97.7%
90%
91.5%
95.2%
99.2%
86.6%
49.1%
53.8%
17.9%
GPT-4.5 Preview
OpenAI
79.2%
77.9%
98.4%
96.2%
96.9%
66.4%
94.5%
99.2%
17.7%
Grok 3
xAI
86.4%
80.5%
95.4%
93.8%
75.2%
79.5%
79.5%
58.7%
56.6%
17.6%
GPT-4o
OpenAI
45.1%
31.7%
0%
81.3%
90.8%
90%
70.4%
77.2%
85.8%
29.2%
82.1%
17.1%
deepseek-v3-0324
DeepSeek
68.5%
96.1%
86.9%
86.2%
97.6%
93.7%
97.6%
52.4%
17%
qwen2.5-max
Qwen
66.2%
93%
93.8%
95.4%
76%
90.6%
94.5%
38%
26.4%
16.8%
GPT-5.1
OpenAI
84.2%
15.6%
80.8%
15.6%
67.9%
0%
63.5%
7.5%
0%
83.3%
50%
91.7%
69.2%
30%
16.5%
mimo-v2.5
Unknown
79.2%
81.3%
75%
81.3%
96.7%
36.3%
29.2%
4%
44.4%
76%
56%
16.5%
o3 mini
OpenAI
52.7%
95.3%
84.6%
88.5%
83.2%
96.1%
90.6%
58.7%
7.1%
16.4%
DeepSeek Reasoner
DeepSeek
67.8%
94.5%
89.2%
93.1%
84%
96.9%
81.1%
46.1%
16.3%
o1 Preview
OpenAI
60.9%
77.3%
100%
100%
94.4%
83.5%
70.1%
56.8%
16.1%
glm-5.1
Zhipu
97.8%
96.9%
96.9%
98.9%
68.4%
45.8%
64%
11.1%
4%
48%
15.8%
qwen3.6-plus
Qwen
91.2%
85.9%
85.9%
97.4%
34.9%
58.3%
80%
72%
12%
15.4%
Claude Sonnet 3.5
Anthropic
43.2%
42.3%
91.4%
92.3%
94.6%
61.6%
85.8%
96.9%
15.2%
DeepSeek Chat
DeepSeek
85.5%
32.8%
32.8%
71.1%
63.1%
54.6%
48.8%
63.8%
76.4%
73.4%
15.1%
deepseek-v3
DeepSeek
56.2%
89.8%
75.4%
80%
72.8%
89.8%
91.3%
38%
14.8%
gemma-4-31b
Google
91.8%
46.9%
89.4%
46.9%
31.6%
86%
76.4%
8.3%
32%
0%
84%
14.8%
Claude Opus 4.5
Anthropic
93.4%
87.5%
87.5%
57.1%
73.7%
92.3%
13.7%
80%
14.6%
mimo-v2.5-pro
Unknown
97.2%
89.1%
89.1%
80.1%
27.4%
12.5%
12%
66.7%
36%
64%
14.3%
llama-4-maverick-17b-128e-instruct
Meta
47%
44.2%
76.6%
83.1%
86.9%
81.6%
75.6%
76.4%
14.3%
Claude Sonnet 4.5
Anthropic
88.6%
50%
60.6%
50%
42.9%
57.9%
83%
17.9%
63.3%
13.3%
21.4%
0%
21.4%
14.3%
GPT-4 Turbo
OpenAI
39.1%
30.8%
70.3%
80%
82.3%
52.8%
70.1%
73.2%
29.2%
33.5%
14%
GPT-4o mini
OpenAI
46.4%
28.8%
68%
66.2%
55.4%
55.2%
63.8%
70.1%
23.2%
72.6%
13.7%
glm-5
Zhipu
92.7%
75%
75%
97.4%
69.8%
40%
24%
72%
13.6%
qwen3.6-max-preview
Qwen
91.8%
90.6%
90.6%
54.2%
68%
94.4%
56%
0%
13.6%
GPT-5.2
OpenAI
77.9%
34.4%
79.8%
34.4%
53.6%
0%
77.1%
0.5%
100%
86.7%
13.6%
deepseek-r1-local-2
DeepSeek
93.8%
87.7%
90.8%
96.8%
97.6%
76.4%
13.6%
Claude Haiku 3.5
Anthropic
37.9%
36.5%
78.1%
71.5%
66.9%
50.4%
74%
81.1%
46.1%
13.6%
Claude Opus 4
Anthropic
64.7%
50%
58.7%
50%
7.1%
75.6%
56.1%
53.3%
33.3%
42.9%
42.9%
13.4%
Claude Opus 3
Anthropic
39.1%
22.1%
60.2%
86.2%
92.3%
44.8%
54.3%
62.2%
43.9%
12.6%
o3
OpenAI
76.7%
73.1%
35.7%
84.9%
19.3%
100%
36.4%
58.3%
20%
12.6%
o1 mini
OpenAI
52.1%
75.8%
83.8%
83.1%
67.2%
91.3%
46.5%
12.5%
dreamina-seedance-2.0-720p
Unknown
100%
100%
100%
98.6%
98.4%
12.4%
mimo-v2-pro
Unknown
89.6%
73.4%
73.4%
96.7%
31.1%
70.8%
36%
8%
16%
12.4%
sonar-pro
Unknown
89.1%
96.9%
99.2%
32%
87.4%
89.8%
12.4%
step-2-16k-202411
StepFun
75%
94.6%
79.2%
99.2%
61.4%
81.1%
12.3%
gemma-3-27b-it
Google
63.1%
49%
63.3%
62.3%
65.4%
64%
59.8%
62.2%
12.2%
Gemini 3 Flash
Google
94.3%
57.8%
90.4%
57.8%
75%
15.8%
79%
18.9%
12.2%
Claude Opus 4.1
Anthropic
79.2%
50%
50%
39.3%
80.1%
16.5%
53.3%
33.3%
42.9%
42.9%
12.2%
codegen3_5k-qwen2.5-72b-instruct-2-chk-50
Unknown
85.9%
70.8%
63.1%
100%
75.6%
91.3%
12.2%
DeepSeek R1 Distill Llama 70B
DeepSeek
79.7%
48.5%
49.2%
57.6%
92.9%
51.2%
38%
63.7%
12%
mistral-large-2407
Mistral
40.7%
73.4%
73.8%
69.2%
53.6%
70.9%
74%
23.2%
12%
happyhorse-1.0
Unknown
97.4%
97.4%
80%
100%
100%
11.9%
Qwen2.5-72B-Instruct-Turbo
Qwen
85.2%
78.5%
62.3%
80%
73.2%
92.9%
11.8%
Grok 2
xAI
50.5%
66.4%
79.2%
76.9%
60%
63.8%
66.1%
11.6%
minimax-m2.7
Unknown
72.9%
68.8%
68.8%
97.4%
58.5%
25%
8%
16.7%
16%
28%
11.5%
hunyuan-turbos-20250313
Tencent
71.9%
76.9%
76.2%
91.2%
80.3%
59.1%
11.4%
Meta-Llama-3.1-405B-Instruct-Turbo
Meta
67.2%
88.5%
85.4%
77.6%
63.8%
69.3%
11.3%
GPT-4
OpenAI
27.1%
56.3%
80.8%
89.2%
64.8%
48%
62.2%
23.2%
11.3%
mistral-large-2411
Mistral
40.1%
73.4%
65.4%
70.8%
56%
70.9%
74%
11.3%
Grok 3 mini
xAI
63.4%
82.8%
56.2%
63.8%
36%
94.5%
51.2%
11.2%
muse-spark
Unknown
98.7%
92.2%
99%
92.2%
63.2%
11.1%
wbot-4:347b_no_s
Unknown
92.2%
50.8%
53.8%
65.6%
87.4%
95.3%
11.1%
qwen2.5-coder-32b-instruct
Qwen
35.3%
86.7%
58.5%
43.8%
50.4%
83.5%
86.6%
11.1%
glm-4.7
Zhipu
90.2%
79.7%
79.7%
77.1%
74.5%
43.3%
11.1%
Grok 2 mini
xAI
43.2%
57.8%
69.2%
68.5%
84.8%
61.4%
51.2%
10.9%
sonar
Unknown
54.7%
91.5%
80.8%
90.4%
70.9%
43.3%
10.8%
Grok Beta
xAI
68.8%
73.1%
81.5%
59.2%
63.8%
74%
10.5%
Qwen3.5 397B A17B
Qwen
91.2%
56.3%
83.7%
56.3%
88.9%
40.1%
10.4%
learnlm-1.5-pro-experimental
Google
72.7%
66.9%
71.5%
58.4%
59.8%
81.1%
10.3%
dracarys2-72b-instruct
Unknown
87.5%
50%
56.9%
42.4%
78.7%
92.9%
10.2%
amazon-nova-pro-v1.0
Amazon
38.5%
13.5%
58.6%
60.8%
64.6%
56.8%
63.8%
51.2%
10.2%
coding-meta-llama-3.1-70b-instruct-chk-50
Unknown
49.2%
77.7%
74.6%
98.4%
56.7%
46.5%
10.1%
coding2-amcfull-apifull-mmlu12k-meta-llama-3.1-70b-instruct-chk-150
Unknown
52.3%
76.2%
73.8%
92%
48%
59.1%
10%
Grok 4
xAI
77.3%
56.7%
28.6%
91.1%
10.8%
12%
40%
23.3%
0%
14.3%
7.1%
40%
10%
llama-3.3-70b-instruct-turbo
Meta
53.9%
70%
73.1%
79.2%
54.3%
56.7%
9.7%
gemma-2-27b-it
Google
35.6%
50.8%
68.5%
58.5%
63.2%
51.2%
56.7%
9.6%
lcb-math-qwen2-72b-instructv3-merged-50
Unknown
61.7%
57.7%
50.8%
93.6%
51.2%
66.9%
9.5%
Meta-Llama-3.1-70B-Instruct-Turbo
Meta
47.7%
82.3%
70%
82.4%
47.2%
50.4%
9.5%
deepseek-coder
DeepSeek
65.6%
55.4%
56.2%
74.4%
61.4%
66.9%
9.5%
GPT Image 2
OpenAI
100%
100%
100%
75%
9.4%
Gemini 3.1 Pro
Google
82.1%
20%
90.9%
41.7%
40%
0%
100%
9.4%
kimi-k2.5-thinking
Unknown
92.7%
71.9%
88.5%
71.9%
42.1%
9.2%
Qwen3.5 122B A10B
Qwen
81.7%
40.6%
76%
40.6%
80.1%
45.8%
9.1%
minimax-m2.5
Unknown
63.4%
48.4%
48.4%
91.1%
64.2%
20.8%
28%
9.1%
acm_rewrite_qwen2-72B-Chat
Unknown
62.5%
51.5%
47.7%
78.4%
53.5%
68.5%
9.1%
Hermes 3 - Llama-3.1 70B
Unknown
44.5%
60%
78.5%
12.8%
40.9%
46.5%
16.6%
55.7%
8.9%
llama-3.1-nemotron-70b-instruct
Meta
45.1%
46.9%
53.1%
60%
54.4%
45.7%
46.5%
8.8%
deepseek-r1-local
DeepSeek
100%
63.8%
83.8%
98.4%
8.7%
qwen2-72b-instruct
Qwen
30%
43%
52.3%
45.4%
96%
26.8%
51.2%
8.6%
dracarys2-llama-3.1-70b-instruct
Unknown
52.3%
64.6%
72.3%
46.4%
48%
59.1%
8.6%
Nemotron 3 Nano Omni 30B A3B Reasoning
Unknown
50.2%
78.8%
4.2%
12%
0%
100%
92%
8.4%
phi-4
Unknown
31.2%
43.8%
39.2%
46.2%
19.2%
45.7%
43.3%
11.8%
54.2%
8.4%
Claude Sonnet 3
Anthropic
31.5%
12.5%
36.7%
56.9%
66.2%
40.8%
40.9%
36.2%
11.8%
8.3%
Qwen3.5 27B
Qwen
75.7%
35.9%
76%
35.9%
83%
21.2%
8.2%
Gemma 4 26B A4B
Google
89.9%
39.1%
84.6%
39.1%
70.8%
8.1%
sky-t1-32b-preview
Unknown
83.6%
28.5%
33.1%
15.2%
80.3%
81.1%
8%
Claude Sonnet 4
Anthropic
58%
60.6%
75.6%
69.3%
13.3%
21.4%
0%
21.4%
8%
GPT-OSS 120B
OpenAI
65.3%
56.8%
83%
10%
8.3%
92.3%
7.9%
mixtral-8x22b-instruct-v0.1
Unknown
23.3%
46.1%
40.8%
39.2%
71.2%
33.9%
55.9%
7.8%
glm-4.6
Zhipu
90.9%
37.5%
37.5%
68.6%
62.3%
13.3%
7.8%
Mistral Small (Feb)
Mistral
28.9%
45.4%
32.3%
49.6%
21.3%
31.5%
9.6%
91.5%
7.8%
kimi-k2.5-instant
Unknown
83.6%
68.8%
82.7%
68.8%
7.6%
deepseek-v4-pro
DeepSeek
94.3%
97.4%
59.9%
0%
0%
0%
28%
24%
7.6%
mistral-small-2501
Mistral
50%
46.9%
52.3%
44%
48%
56.7%
7.4%
pixverse-v5.6
Unknown
57.9%
60.5%
87.5%
88.9%
7.4%
mistral-large-2402
Mistral
26.5%
38.3%
43.1%
44.6%
60.8%
40.9%
40.2%
7.4%
mistral-small-2503
Mistral
51.6%
44.6%
53.1%
36.8%
56.7%
51.2%
7.3%
minimax-m2
Unknown
59.3%
28.1%
28.1%
80.1%
63.2%
33.3%
7.3%
Meta-Llama-3-70B-Instruct
Meta
28.1%
59.2%
59.2%
80.8%
40.2%
23.6%
7.3%
Grok 4.1 Fast
xAI
76.7%
12.5%
63.5%
12.5%
50%
56.8%
15.1%
7.2%
GPT-4.1
OpenAI
70%
70.2%
60.9%
83.5%
7.1%
GPT-5.3 Codex
OpenAI
45.3%
45.3%
94.8%
1.4%
96.7%
7.1%
kling-2.5-turbo-1080p
Unknown
52.6%
55.3%
79.2%
92.1%
7%
Claude Haiku 3
Anthropic
29.3%
9.6%
33.6%
49.2%
46.9%
24.8%
29.1%
36.2%
19.6%
7%
Qwen2.5-7B-Instruct-Turbo
Qwen
59.4%
36.9%
27.7%
33.6%
59.1%
59.1%
6.9%
o4 mini
OpenAI
60.9%
65.4%
75.6%
6.6%
62.5%
6.8%
Kimi K2.5
Unknown
83%
30.2%
16.7%
24%
22.2%
20%
8%
66.7%
6.8%
glm-5v-turbo
Zhipu
78.8%
37.5%
20%
22.2%
80%
32%
6.8%
qwen2-math-72b-instruct
Qwen
64.1%
29.2%
31.5%
18.4%
63.8%
62.2%
6.7%
GPT-4.1 mini
OpenAI
58.7%
66.3%
54.2%
86.8%
6.7%
vidu-q3-pro
Unknown
78.9%
93.1%
93.7%
6.6%
amazon-nova-lite-v1.0
Amazon
34.4%
13.5%
39.1%
36.2%
41.5%
25.6%
29.1%
43.3%
6.6%
deepseek-v4-pro-thinking
DeepSeek
93.7%
84.4%
84.4%
6.6%
GPT-OSS 20B
OpenAI
46.7%
50.2%
88.2%
0%
76.9%
6.6%
deepseek-v4-flash
DeepSeek
88%
94.8%
75.9%
6.5%
qwq-32b-preview
Unknown
23.3%
55.5%
23.1%
34.6%
3.2%
82.7%
31.5%
6.3%
GPT-3.5 Turbo
OpenAI
14.2%
37.5%
38.5%
38.5%
39.2%
34.6%
40.9%
9.6%
6.3%
trinity-large-thinking
Unknown
60.6%
18.8%
18.8%
73.4%
81.1%
6.3%
Grok Imagine Video
xAI
60%
94.4%
96.8%
6.3%
kling-2.6-pro
Unknown
50%
57.9%
72.2%
69.8%
6.2%
veo-3
Unknown
63.2%
44.7%
84.7%
54%
6.2%
command-r-plus-08-2024
Cohere
35%
24.2%
53.1%
50%
34.4%
26%
23.6%
6.2%
Mistral Large 3
Mistral
88.6%
10.9%
10.9%
54.2%
81.1%
6.1%
DeepSeek-R1-Distill-Qwen-32B
DeepSeek
48.4%
34.6%
43.1%
14.4%
85%
19.7%
6.1%
azerogpt
Unknown
44.5%
37.7%
40.8%
32.8%
40.9%
46.5%
6.1%
step-3.5-flash
StepFun
75.4%
84.9%
79.2%
6%
qwen3-vl-235b-a22b-instruct
Qwen
83.6%
72.1%
50.2%
33%
6%
gemma-2-9b-it
Google
30.6%
32.8%
43.8%
40%
30.4%
29.1%
31.5%
6%
Gemini 2.5 Flash
Google
76%
70.2%
50.2%
9%
6.7%
25%
5.9%
Qwen3.5 35B A3B
Qwen
73.5%
20.3%
20.3%
70.8%
50.5%
5.9%
Phi-3-small-128k-instruct
Unknown
35.2%
30.8%
26.9%
71.2%
34.6%
36.2%
5.9%
GPT Image 1.5
OpenAI
50.8%
8.7%
75%
100%
5.9%
mimo-v2-flash (thinking)
Unknown
72.9%
26.6%
26.6%
68.6%
35.4%
5.8%
qwen1.5-72b-chat
Qwen
24.9%
32%
25.4%
17.7%
69.6%
22.8%
34.6%
5.7%
nvidia-nemotron-3-super-120b-a12b
Unknown
69.4%
80.1%
74.5%
5.6%
runway-gen-4.5
Unknown
55.3%
86.1%
81%
5.6%
Command-R+ (Apr)
Cohere
25.8%
47.7%
35.4%
48%
32.3%
23.6%
5.9%
5.5%
Qwen3-Coder 480B A35B
Qwen
62.1%
23.4%
23.4%
58.7%
30.7%
16.7%
5.4%
command-r-08-2024
Cohere
27.4%
21.1%
41.5%
36.2%
43.2%
18.1%
26%
5.3%
deepseek-r1-0528
DeepSeek
87.7%
93.3%
0%
30.8%
0%
0%
5.3%
hailuo-2.3
Unknown
36.8%
50%
63.9%
58.7%
5.2%
open-mixtral-8x22b
Unknown
39.8%
32.3%
36.9%
13.6%
44.9%
40.9%
5.2%
Phi-3-medium-128k-instruct
Unknown
31.3%
26.9%
20%
68%
32.3%
29.9%
5.2%
mimo-v2-omni
Unknown
67.3%
92.3%
47.2%
5.2%
open-mistral-nemo
Unknown
41.4%
24.6%
23.1%
35.2%
34.6%
43.3%
5.1%
MiniMax-M2.1
Unknown
86%
65.6%
50%
5%
DeepSeek V3.2 Exp
DeepSeek
85.5%
25%
25%
64.9%
5%
Grok 4 Fast
xAI
73.5%
3.1%
3.1%
46.4%
54.2%
19.8%
5%
ray-3
Unknown
42.1%
28.9%
72.2%
52.4%
4.9%
minimax-m2.1-preview
Unknown
71.6%
60.9%
60.9%
4.8%
Reflection-Llama-3.1-70B
Unknown
20.3%
46.2%
57.7%
26.4%
27.6%
12.6%
4.8%
hailuo-02-pro
Unknown
36.8%
31.6%
54.2%
66.7%
4.7%
GPT-4.1 nano
OpenAI
46.1%
26.9%
23.2%
92.5%
4.7%
mistral-small-2409
Mistral
35.2%
33.1%
42.3%
7.2%
34.6%
36.2%
4.7%
Kling 3.0 1080p (Pro)
Unknown
97.2%
90.5%
4.7%
ERNIE 5.1
Baidu
98.1%
89.3%
4.7%
dola-seed-2.0-pro
Unknown
95.6%
90.4%
4.6%
Phi-3.5-MoE-instruct
Unknown
30.5%
27.7%
28.5%
37.6%
34.6%
26%
4.6%
qwen3-235b-a22b-instruct-2507
Qwen
81.7%
58.7%
44.3%
4.6%
veo-3.1-audio-1080p
Unknown
94.7%
89.5%
4.6%
gpt-image-1.5-high-fidelity
OpenAI
94.9%
89.1%
4.6%
KAT-Coder-Pro-V1
Unknown
21.9%
21.9%
80.1%
59%
4.6%
PixVerse V6
Unknown
87.5%
95.2%
4.6%
Kling 3.0 Omni 1080p (Pro)
Unknown
94.4%
87.3%
4.5%
veo-3.1-audio
Unknown
89.5%
92.1%
4.5%
glm-4.6v
Zhipu
68.5%
52.9%
41.3%
16%
4.5%
grok-imagine-image-quality
xAI
91.5%
87%
4.5%
qwen3-next-80b-a3b-instruct
Qwen
80.8%
49.1%
48.6%
4.5%
glm-4.5
Zhipu
88%
60.9%
28.3%
4.4%
hailuo-02-standard
Unknown
31.6%
23.7%
62.5%
58.7%
4.4%
grok-imagine-video-720p
xAI
81.6%
94.7%
4.4%
gpt-5.5-instant
OpenAI
80.8%
95.2%
0%
4.4%
glm-4.7-flash
Zhipu
61.5%
52.4%
61.3%
4.4%
Kimi K2 Thinking
Unknown
89.3%
39.2%
46.7%
4.4%
Meta-Llama-3-8B-Instruct
Meta
25%
30%
29.2%
41.6%
22.8%
26%
4.4%
uni-1.1-max
Unknown
89.8%
84.8%
4.4%
mercury-2
Unknown
62.8%
4.7%
4.7%
75.6%
23.6%
4.3%
veo-3.1-fast-audio
Unknown
84.2%
86.8%
4.3%
deepseek-v3.2-thinking
DeepSeek
81.7%
43.8%
43.8%
4.2%
Grok 3 mini Reasoning (high)
xAI
73.4%
95.3%
4.2%
phi-3-small-8k-instruct
Unknown
17.4%
22.7%
33.8%
25.4%
23.2%
19.7%
26%
4.2%
Kling 3.0 Omni 720p (Standard)
Unknown
90.3%
77.8%
4.2%
Qwen3.5 9B
Qwen
73.4%
93.9%
4.2%
Llama 4 Maverick
Meta
43.9%
80.2%
0%
0%
23.1%
0%
20%
4.2%
qwen1.5-110b-chat
Qwen
26.2%
29.7%
16.2%
20.8%
20%
22.8%
31.5%
4.2%
uni-1.1
Unknown
88.1%
78.3%
4.2%
veo-3.1-fast-audio-1080p
Unknown
84.2%
81.6%
4.1%
Grok Imagine Image
xAI
84.7%
80.4%
4.1%
amazon-nova-micro-v1.0
Amazon
30.9%
26.6%
21.5%
30%
4.8%
21.3%
29.9%
4.1%
Step 3.5 Flash
StepFun
84.9%
79.7%
4.1%
Qwen3.5 4B
Qwen
63.5%
99.1%
4.1%
Meta-Llama-3.1-8B-Instruct-Turbo
Meta
23.4%
35.4%
30.8%
20.8%
34.6%
17.3%
4.1%
phi-3-medium-4k-instruct
Unknown
21.8%
27.3%
26.2%
22.3%
9.6%
29.1%
26%
4.1%
hunyuan-hy3-preview
Tencent
77.3%
42.2%
42.2%
4%
Qwen2-7B-Instruct
Qwen
42.2%
9.2%
14.6%
8%
51.2%
36.2%
4%
Kling 3.0 720p (Standard)
Unknown
83.3%
77.8%
4%
Grok Imagine Image Pro
xAI
78%
82.6%
4%
Qwen3 Max
Qwen
77.9%
60.9%
21.7%
4%
Veo 3.1 Fast
Unknown
76.4%
82.5%
4%
sora-2-pro
Unknown
92.1%
66.7%
4%
Smaug-Qwen2-72B-Instruct
Unknown
57%
25.2%
76.4%
4%
Grok 4.20 0309
xAI
68.6%
89.2%
3.9%
command-a-03-2025
Unknown
55.2%
40%
61.5%
3.9%
Veo 3.1 Lite
Unknown
81.9%
74.6%
3.9%
PixVerse V5.5
Unknown
72.2%
84.1%
3.9%
GLM-5.1
Zhipu
94.1%
60.8%
3.9%
DeepSeek-Coder-V2-Lite-Instruct
DeepSeek
34.4%
13.8%
15.4%
21.6%
34.6%
34.6%
3.9%
Kling 2.6 Standard (January)
Unknown
68.1%
85.7%
3.8%
Phi-3.5-mini-instruct
Unknown
19.5%
19.2%
13.8%
62.4%
18.1%
20.5%
3.8%
DeepSeek V4 Flash
DeepSeek
80.1%
73.1%
3.8%
Qwen3.5 4B
Qwen
54.2%
98.6%
3.8%
glm-4.5-air
Zhipu
70%
54.2%
27.8%
3.8%
p-video
Unknown
47.4%
42.1%
36.1%
25.4%
3.8%
Mistral Medium 3.5
Mistral
86%
64.6%
3.8%
Gemini 2.5 Flash-Lite
Google
66.9%
54.8%
23.2%
5.7%
3.8%
deepseek-v3.1-terminus
DeepSeek
80.8%
67.5%
3.7%
DeepSeek V4 Pro
DeepSeek
86%
61.8%
3.7%
KAT-Coder-Pro V2
Unknown
94.1%
53.3%
3.7%
veo-3-audio
Unknown
73.7%
73.7%
3.7%
veo-3-fast-audio
Unknown
78.9%
68.4%
3.7%
Mistral Medium 3
Mistral
46.1%
71.2%
30%
3.7%
deepseek-v3.1
DeepSeek
81.7%
64.9%
3.7%
Veo 3.1
Unknown
76.4%
68.3%
3.6%
qwen-image-2.0-pro-2026-04-22
Qwen
83.1%
60.9%
3.6%
Qwen3 235B A22B
Qwen
65.6%
6.7%
7.1%
14.3%
50%
3.6%
ling-flash-2.0
Unknown
63.4%
38%
41.5%
3.6%
kling-o1-pro
Unknown
42.1%
20%
80.6%
3.6%
Devstral 2
Mistral
6.3%
6.3%
52.4%
77.4%
3.6%
GLM-5
Zhipu
89.3%
52.8%
3.6%
openhermes-2.5-mistral-7b
Unknown
13.9%
16.4%
20.8%
17.7%
45.6%
15%
12.6%
3.5%
Qwen3 Coder Next
Qwen
64.9%
76.4%
3.5%
kimi-k2-thinking-turbo
Unknown
78.5%
31.3%
31.3%
3.5%
glm-4.5v
Zhipu
56.8%
47.1%
23.2%
12.7%
3.5%
gemma-3-12b-it
Google
56.8%
31.5%
51.5%
3.5%
Mistral Small 4
Mistral
46.1%
92.5%
3.5%
flux-2-max
Unknown
79.7%
58.7%
3.5%
seedream-4.5
Unknown
66.1%
71.7%
3.4%
qwen3-vl-235b-a22b-thinking
Qwen
75.1%
62.5%
3.4%
MiMo-V2-Flash
Unknown
86%
51.4%
3.4%
mistral-medium-2508
Mistral
87.1%
49%
3.4%
PixVerse V5
Unknown
59.7%
76.2%
3.4%
Kimi K2 0905
Unknown
70.8%
37.7%
26.7%
3.4%
MiMo-V2-Omni-0327
Unknown
94.8%
40.1%
3.4%
Qwen3.6 35B A3B
Qwen
92.3%
41.5%
3.3%
Qwen3.5 2B
Qwen
33.2%
100%
3.3%
Magistral Small 1.2
Mistral
43.9%
88.2%
3.3%
Seedance 1.5 pro
Unknown
59.7%
71.4%
3.3%
Command-R (Mar)
Cohere
18%
22.3%
23.8%
27.2%
15%
20.5%
4.1%
3.3%
Qwen3 32B
Qwen
58.7%
33.2%
38.2%
3.3%
seedance-v1.5-pro
Unknown
65.8%
63.2%
3.2%
Ministral 3 14B
Mistral
38%
90.6%
3.2%
Magistral Medium 1.2
Mistral
63.5%
65.1%
3.2%
Mistral Small 3.2
Mistral
33.2%
95.3%
3.2%
mathstral-7B-v0.1
Unknown
17.2%
23.8%
26.2%
28.8%
19.7%
12.6%
3.2%
sora-2
Unknown
71.1%
56.9%
3.2%
hunyuan-vision-1.5-thinking
Tencent
76%
51.9%
3.2%
Ministral 3 8B
Mistral
33.2%
94.3%
3.2%
NVIDIA Nemotron Nano 9B V2
Unknown
33.2%
94.3%
3.2%
Qwen3.5 Omni Plus
Qwen
86%
41.5%
3.2%
seedream-4-2k
Unknown
64.4%
63%
3.2%
ring-flash-2.0
Unknown
55.2%
29.2%
41.5%
3.1%
Devstral Small
Mistral
33.2%
91.5%
3.1%
flux-2-pro
Unknown
74.6%
50%
3.1%
seedream-5.0-lite
Unknown
57.6%
65.2%
3.1%
pixtral-large-2411
Unknown
27.9%
29.2%
65.6%
3.1%
Qwen3.6 Max Preview
Qwen
99.3%
22.6%
3%
Mistral Small 3.1
Mistral
29.2%
92.5%
3%
Qwen3.5 Omni Flash
Qwen
60.9%
60.4%
3%
Llama 3.1 Nemotron 70B
Meta
23.2%
97.6%
3%
Devstral Small 2
Mistral
46.1%
74.5%
3%
Wan 2.5 Preview
Unknown
55.6%
65.1%
3%
mistral-7b-instruct-v0.2
Mistral
13.6%
15.6%
15.4%
11.5%
37.6%
14.2%
12.6%
3%
olmo-2-1124-13b-instruct
Unknown
10.9%
20%
16.9%
47.2%
7.9%
17.3%
3%
Mistral Medium 3.1
Mistral
50.2%
69.8%
3%
Qwen3.6 27B
Qwen
95.9%
23.6%
3%
Wan 2.6
Unknown
68.1%
50.8%
3%
Devstral Medium
Mistral
46.1%
72.2%
3%
mixtral-8x7b-instruct-v0.1
Unknown
20.8%
11.7%
17.7%
21.5%
28%
11.8%
6.3%
2.9%
Hy3-preview
Unknown
91.1%
25.9%
2.9%
wan2.7-image-pro
Unknown
44.1%
71.7%
2.9%
starling-lm-7b-beta
Unknown
20.8%
21.9%
7.7%
13.1%
4%
27.6%
20.5%
2.9%
mistral-medium-2505
Mistral
67.2%
47.1%
2.9%
wan2.7-image
Unknown
42.4%
71.7%
2.9%
Llama 4 Scout
Meta
29.2%
84.9%
2.9%
Mistral Small 3
Mistral
23.2%
90.6%
2.8%
Ministral 3 3B
Mistral
16.6%
97.2%
2.8%
flux-2-flex
Unknown
74.6%
39.1%
2.8%
Qwen3.6 35B A3B
Qwen
73.4%
39.6%
2.8%
Nova 2.0 Lite
Unknown
43.9%
68.9%
2.8%
Seedance 1.0
Unknown
52.8%
58.7%
2.8%
Qwen3.5 0.8B
Qwen
11.8%
99.5%
2.8%
Grok Code Fast
xAI
1.6%
1.6%
67.5%
20.3%
20%
2.8%
qwen-image-2.0-2026-03-03
Qwen
55.9%
54.3%
2.8%
LFM2 24B A2B
Unknown
11.8%
97.6%
2.7%
Grok 4.1
xAI
90.5%
9.4%
9.4%
2.7%
Qwen3.5 Plus
Qwen
60%
44%
4%
2.7%
veo-3-fast
Unknown
60.5%
47.4%
2.7%
phi-3-mini-4k-instruct
Unknown
11%
18.8%
18.5%
9.2%
12%
17.3%
20.5%
2.7%
Granite 4.1 8B
Unknown
19.6%
87.3%
2.7%
Llama Nemotron Super 49B v1.5
Meta
33.2%
73.6%
2.7%
step-1o-turbo-202506
StepFun
57.4%
49%
2.7%
Qwen3.6 27B
Qwen
83%
23.1%
2.7%
Llama 3.1 8B
Meta
19.6%
85.4%
2.6%
mistral-7b-instruct
Mistral
5%
4.1%
95.3%
2.6%
flux-2-dev
Unknown
69.5%
34.8%
2.6%
Hy3-preview
Unknown
77.1%
26.9%
2.6%
qwen2.5-72b-instruct
Qwen
41%
38%
24.5%
2.6%
Nova Lite
Unknown
23.2%
80.2%
2.6%
Claude 4 Opus
Anthropic
86%
17%
2.6%
Qwen3.5 Flash
Qwen
74.1%
14.1%
14.1%
2.6%
step-3
StepFun
59.9%
42.3%
2.6%
NVIDIA Nemotron Nano 9B V2
Unknown
23.2%
78.3%
2.5%
Phi-4 Mini
Unknown
5.9%
95.3%
2.5%
Mistral Small (Sep)
Mistral
11.8%
89.2%
2.5%
mistral-medium
Mistral
24.3%
9.6%
67%
2.5%
Qwen3 235B A22B 2507
Qwen
68.6%
32.1%
2.5%
Kling 2.1 Master
Unknown
51.4%
49.2%
2.5%
mistral-small-2506
Mistral
58.4%
41.3%
2.5%
Llama 3.2 11B (Vision)
Meta
9.6%
90.1%
2.5%
veo-2
Unknown
21.1%
10.5%
47.2%
20.6%
2.5%
Claude 4 Sonnet
Anthropic
86%
12.3%
2.5%
NVIDIA Nemotron Nano 12B v2 VL
Unknown
11.8%
86.3%
2.5%
qwen3.5-max-preview
Qwen
98.1%
2.5%
chatgpt-image-latest-high-fidelity (20251216)
Unknown
97.8%
2.4%
Kimi K2
Unknown
60.9%
36.8%
2.4%
Qwen3 30B A3B 2507
Qwen
52.4%
44.3%
2.4%
vidu-q2-turbo
Unknown
39.5%
57.1%
2.4%
Llama 3.3 70B
Meta
29.2%
67%
2.4%
Llama-3.1-Nemotron-70B-Instruct-HF
Meta
42.3%
37.7%
16%
2.4%
Qwen3 Max Thinking
Qwen
73.4%
22.2%
2.4%
Hermes 4 70B
Unknown
23.2%
71.7%
2.4%
LTX-2 Pro Open Weights
Unknown
50%
44.4%
2.4%
Sarvam 105B (high)
Unknown
43.9%
50.5%
2.4%
Nova Micro
Unknown
11.8%
82.5%
2.4%
amazon-nova-experimental-chat-26-02-10
Amazon
94.3%
2.4%
ernie-5.0-0110
Unknown
94.3%
2.4%
Qwen3 VL 235B A22B
Qwen
64.9%
28.8%
2.3%
Llama 3.2 90B (Vision)
Meta
19.6%
74.1%
2.3%
Claude 4.5 Haiku
Anthropic
83%
10.4%
2.3%
Qwen3 Omni 30B A3B
Qwen
38%
54.7%
2.3%
o3-pro
OpenAI
89.3%
2.8%
2.3%
ernie-5.0-preview-1203
Unknown
91.8%
2.3%
Nova 2.0 Pro Preview (medium)
Unknown
80.1%
11.3%
2.3%
Bach-1.0 Preview
Unknown
90.3%
2.3%
Qwen3 VL 30B A3B
Qwen
38%
51.9%
2.2%
LTX-2 Fast Open Weights
Unknown
48.6%
41.3%
2.2%
Llama 3.2 3B
Meta
11.8%
77.8%
2.2%
Llama 3.1 405B
Meta
41.3%
48.1%
2.2%
Jamba 1.6 Mini
Unknown
5.9%
83.5%
2.2%
ernie-5.0-preview-1022
Unknown
88.6%
2.2%
Qwen3 30B A3B 2507
Qwen
33.2%
54.7%
2.2%
Seed-OSS-36B-Instruct
Unknown
58.7%
29.2%
2.2%
Pika 2.5
Unknown
40.3%
47.6%
2.2%
Hermes 4 405B
Unknown
43.9%
43.9%
2.2%
Nova 2.0 Lite (high)
Unknown
79%
8.5%
2.2%
longcat-flash-chat-2602-exp
Unknown
87.1%
2.2%
seedance-v1-pro
Unknown
34.2%
52.6%
2.2%
mai-image-2
Unknown
86.4%
2.2%
deepseek-v4-flash-thinking
DeepSeek
86.4%
2.2%
deepseek-v3.2-exp-thinking
DeepSeek
86.1%
2.2%
zephyr-7b-alpha
Unknown
9.5%
13.3%
10%
7.7%
22.4%
15%
7.9%
2.1%
Cogito v2.1
Unknown
85.4%
2.1%
Llama 3.2 1B
Meta
0.7%
84.4%
2.1%
hunyuan-large-vision
Tencent
39.7%
45.2%
2.1%
Qwen3 Coder 30B A3B
Qwen
49.1%
35.8%
2.1%
Jamba 1.7 Large
Unknown
16.6%
67.9%
2.1%
phi-3-mini-128k-instruct
Unknown
6.9%
14.1%
16.9%
10%
10.4%
13.4%
12.6%
2.1%
longcat-flash-chat
Unknown
84.2%
2.1%
grok-imagine-video-480p
xAI
84.2%
2.1%
amazon-nova-experimental-chat-12-10
Amazon
83.3%
2.1%
Mistral Large 2 (Nov)
Mistral
33.2%
50%
2.1%
Jamba 1.6 Large
Unknown
16.6%
66.5%
2.1%
gemma-1.1-7b-it
Google
14.2%
10.2%
12.3%
16.2%
17.6%
12.6%
0%
2.1%
mistral-small-3.1-24b-instruct-2503
Mistral
42.9%
39.4%
2.1%
Mistral-7B-Instruct-v0.3
Mistral
12.5%
14.6%
19.2%
16.8%
11%
7.9%
2.1%
seedream-4-high-res-fal
Unknown
49.2%
32.6%
2%
deepseek-v3.1-terminus-thinking
DeepSeek
81.7%
2%
llama-4-scout-17b-16e-instruct
Meta
43.2%
38.5%
2%
Qwen3 VL 8B
Qwen
41.3%
40.1%
2%
seedream-4-fal
Unknown
52.5%
28.3%
2%
LTX-2.3 Fast Open Weights
Unknown
45.8%
34.9%
2%
Kling 2.0
Unknown
41.7%
38.1%
2%
reve-v1.5
Unknown
79.7%
2%
vidu-q2-pro
Unknown
23.7%
55.6%
2%
amazon-nova-experimental-chat-11-10
Amazon
79.2%
2%
deepseek-v3.1-thinking
DeepSeek
79.2%
2%
Sarvam 30B (high)
Unknown
19.6%
59.4%
2%
molmo-2-8b
Unknown
48.9%
29.8%
2%
qwen3-235b-a22b-thinking-2507
Qwen
78.5%
2%
Nova 2.0 Lite (medium)
Unknown
68.6%
9.9%
2%
LTX-2.3 Pro Open Weights
Unknown
44.4%
33.3%
1.9%
Qwen3 VL 8B
Qwen
29.2%
48.6%
1.9%
zephyr-7b-beta
Unknown
10.4%
8.6%
8.5%
3.8%
31.2%
9.4%
5.5%
1.9%
vicuna-7b-v1.5-16k
Unknown
5.5%
13.1%
8.5%
40%
2.4%
7.9%
1.9%
gemma-3-4b-it
Google
47.9%
4.6%
24.6%
1.9%
Ling-2.6-1T
Unknown
77.1%
1.9%
Llama 3.1 70B
Meta
19.6%
57.1%
1.9%
kling-v3-pro
Unknown
76.3%
1.9%
Nova Premier
Unknown
46.1%
29.7%
1.9%
Qwen3 VL 32B
Qwen
41.3%
34.4%
1.9%
Command A
Unknown
23.2%
51.9%
1.9%
Nano Banana Pro (Gemini 3 Pro Image)
Unknown
25%
50%
1.9%
hunyuan-t1-20250711
Tencent
74.8%
1.9%
LongCat Flash Lite
Unknown
56.8%
17.5%
1.9%
Qwen3 Omni 30B A3B
Qwen
16.6%
57.5%
1.9%
amazon-nova-experimental-chat-26-01-10
Amazon
74.1%
1.9%
wan2.6-t2v
Unknown
73.7%
1.8%
ernie-5.0-preview-1220
Unknown
73.1%
1.8%
TeleVideo 2.0
Unknown
73%
1.8%
Qwen3 235B
Qwen
41.3%
31.6%
1.8%
amazon-nova-experimental-chat-10-20
Amazon
71.9%
1.8%
mai-1-preview
Unknown
71.9%
1.8%
qwen3-235b-a22b-no-thinking
Qwen
71.9%
1.8%
Llama Nemotron Ultra
Meta
33.2%
38.2%
1.8%
hunyuan-image-3.0
Tencent
71.2%
1.8%
wan2.5-i2v-preview
Unknown
71.1%
1.8%
Qwen3 30B
Qwen
23.2%
47.6%
1.8%
flux-2-klein-9b
Unknown
33.9%
37%
1.8%
Kling O1 Standard (January)
Unknown
70.8%
1.8%
qwen3-30b-a3b-instruct-2507
Qwen
70.7%
1.8%
DeepSeek-V2-Lite-Chat
DeepSeek
7.8%
11.5%
12.3%
24%
7.1%
7.9%
1.8%
Gemma 3n E4B
Google
0.7%
69.8%
1.8%
olmo-3.1-32b-think
Unknown
41.3%
29.2%
1.8%
Qwen3 30B
Qwen
33.2%
37.3%
1.8%
Nova 2.0 Lite (low)
Unknown
58.7%
11.8%
1.8%
kimi-k2-0905-preview
Unknown
69.7%
1.7%
hunyuan-image-3.0-instruct
Tencent
69.6%
1.7%
NVIDIA Nemotron 3 Nano
Unknown
23.2%
46.2%
1.7%
hunyuan-turbos-20250416
Tencent
69.1%
1.7%
Nova 2.0 Pro Preview
Unknown
54.2%
14.6%
1.7%
wan2.5-t2v-preview
Unknown
68.4%
1.7%
imagen-ultra-4.0-generate-001
Unknown
67.8%
1.7%
Seedance 1.0 Mini
Unknown
37.5%
30.2%
1.7%
qwen1.5-7b-chat
Qwen
12.6%
9.4%
6.9%
6.2%
11.2%
8.7%
12.6%
1.7%
kimi-k2-0711-preview
Unknown
67.5%
1.7%
PixVerse V4.5
Unknown
38.9%
28.6%
1.7%
Qwen3 4B
Qwen
19.6%
46.7%
1.7%
qwen3-next-80b-a3b-thinking
Qwen
66.2%
1.7%
Qwen3 8B
Qwen
16.6%
49.5%
1.7%
wan2.6-i2v
Unknown
65.8%
1.6%
Veo 3 Fast Preview
Unknown
65.3%
1.6%
Wan 2.2 A14B Open Weights
Unknown
43.1%
22.2%
1.6%
Apriel-v1.5-15B-Thinker
Unknown
64.9%
1.6%
Apriel-v1.6-15B-Thinker
Unknown
64.9%
1.6%
Nova 2.0 Omni (medium)
Unknown
64.9%
1.6%
amazon-nova-experimental-chat-10-09
Amazon
64.7%
1.6%
Mistral Nemo 12B
Mistral
64.6%
1.6%
hailuo-02-fast
Unknown
18.4%
46%
1.6%
Qwen2.5 Turbo
Qwen
19.6%
44.3%
1.6%
Hailuo 2.3 Fast
Unknown
63.5%
1.6%
nova-2-lite
Unknown
63.4%
1.6%
Qwen3 0.6B
Qwen
0.7%
62.3%
1.6%
wan2.6-t2i
Unknown
62.7%
1.6%
intellect-3
Unknown
62.1%
1.6%
Qwen3 1.7B
Qwen
4.1%
57.5%
1.5%
Reka Flash
Unknown
19.6%
41.5%
1.5%
Ling 2.6 Flash
Unknown
60.9%
1.5%
o1-pro
OpenAI
60.9%
1.5%
nvidia-nemotron-3-nano-30b-a3b-bf16
Unknown
59.9%
1.5%
imagen-4.0-generate-001
Unknown
59.3%
1.5%
qwen-image-2512
Qwen
59.3%
1.5%
minimax-m1
Unknown
59.3%
1.5%
vicuna-7b-v1.5
Unknown
3.1%
10.8%
10.8%
29.6%
4.7%
0%
1.5%
qwen-vl-max-2025-08-13
Qwen
58.7%
1.5%
Vidu Q2
Unknown
58.3%
1.5%
ERNIE 4.5 300B A47B
Unknown
33.2%
25%
1.5%
kling-v2.1-master
Unknown
21.1%
36.8%
1.4%
nvidia-llama-3.3-nemotron-super-49b-v1.5
Unknown
57.7%
1.4%
MiniMax M1 80k
Unknown
56.8%
1.4%
reve-v1.1
Unknown
56.5%
1.4%
trinity-large-preview
Unknown
56.2%
1.4%
Qwen3 14B
Qwen
23.2%
32.1%
1.4%
glm-4-plus-0111
Zhipu
55.2%
1.4%
flux-1-kontext-max
Unknown
35.6%
19.6%
1.4%
qwen-plus-0125
Qwen
54.3%
1.4%
Nova 2.0 Omni (low)
Unknown
54.2%
1.4%
step-2-16k-exp-202412
StepFun
53.9%
1.3%
hunyuan-turbos-20250226
Tencent
53.3%
1.3%
jamba-1.5-large
Unknown
36.3%
16.6%
1.3%
llama-3.1-nemotron-ultra-253b-v1
Meta
52.7%
1.3%
wan2.5-t2i-preview
Unknown
52.5%
1.3%
GPT Image 1 mini
OpenAI
45.8%
6.5%
1.3%
kling-image-o1
Unknown
52.2%
1.3%
Qwen3 30B A3B
Qwen
52.1%
1.3%
olmo-3.1-32b-instruct
Unknown
51.7%
1.3%
hunyuan-turbo-0110
Tencent
51.4%
1.3%
llama-3.3-nemotron-49b-super-v1
Meta
51.1%
1.3%
gemma-3n-e4b-it
Google
50.8%
1.3%
Kling 1.6 Pro
Unknown
23.6%
27%
1.3%
yi-lightning
Unknown
50.2%
1.3%
olmo-3-32b-think
Unknown
49.2%
1.2%
qwen2.5-plus-1127
Qwen
49.2%
1.2%
deepseek-v2.5-1210
DeepSeek
48.6%
1.2%
athene-v2-chat
Unknown
47.9%
1.2%
qwen-image-edit
Qwen
47.8%
1.2%
glm-4-plus
Zhipu
47.6%
1.2%
recraft-v4
Unknown
47.5%
1.2%
hunyuan-video-1.5
Tencent
26.3%
21.1%
1.2%
hunyuan-large-2025-02-10
Tencent
47%
1.2%
Gemma 4 E4B
Google
46.1%
1.2%
HunyuanVideo-1.5 (Fal) Open Weights
Tencent
22.2%
23.8%
1.2%
Marey
Unknown
34.7%
11.1%
1.1%
llama-3.1-405b-instruct-bf16
Meta
45.7%
1.1%
flux-1-kontext-pro
Unknown
30.5%
15.2%
1.1%
reve-v1
Unknown
45.7%
1.1%
llama-3.1-405b-instruct-fp8
Meta
44.2%
1.1%
mercury
Unknown
44.2%
1.1%
qwen-max-0919
Qwen
44.2%
1.1%
open-mixtral-8x7b
Unknown
14.8%
10.2%
18.9%
1.1%
qwen-image-edit-2511
Qwen
43.5%
1.1%
Kling 2.1 Pro
Unknown
42.9%
1.1%
sora
Unknown
7.9%
33.3%
1.6%
1.1%
llama-3.3-70b-instruct
Meta
42.6%
1.1%
hunyuan-standard-2025-02-10
Tencent
42.3%
1.1%
flux-2-klein-4b
Unknown
20.3%
21.7%
1.1%
deepseek-v2.5
DeepSeek
41.6%
1%
Nova 2.0 Omni
Unknown
41.3%
1%
wan2.6-image
Unknown
41.3%
1%
mai-image-1
Unknown
40.7%
1%
athene-70b-0725
Unknown
40.1%
1%
kling-o3-pro
Unknown
40%
1%
z-image-turbo
Unknown
39%
1%
llama-3.1-70b-instruct
Meta
38.8%
1%
llama-3.1-tulu-3-70b
Meta
38.2%
1%
Kling 2.1 Standard
Unknown
38.1%
1%
Claude 3.5 Sonnet (Oct)
Anthropic
38%
1%
runway-gen4
Unknown
18.6%
19%
0.9%
magistral-medium-2506
Mistral
37.5%
0.9%
seedream-3
Unknown
37.3%
0.9%
Phi-4 Multimodal
Unknown
11.8%
25.5%
0.9%
reka-core-20240904
Unknown
37.2%
0.9%
ibm-granite-h-small
Unknown
36.9%
0.9%
Midjourney V1
Unknown
36.5%
0.9%
mistral-small-24b-instruct-2501
Mistral
36%
0.9%
step-1o-vision-32k-highres
StepFun
35.6%
0.9%
qwen2.5-vl-72b-instruct
Qwen
34.6%
0.9%
llama-3.1-nemotron-51b-instruct
Meta
34.4%
0.9%
gemma-2-9b-it-simpo
Google
34.1%
0.9%
Vidu Q1
Unknown
19.4%
14.3%
0.8%
qwen2.5-vl-32b-instruct
Qwen
33.7%
0.8%
Motion 2.0
Unknown
20.8%
12.7%
0.8%
glm-4-0520
Zhipu
33.4%
0.8%
jamba-1.5-mini
Unknown
27.4%
5.9%
0.8%
Gemma 4 E4B
Google
33.2%
0.8%
nemotron-4-340b-instruct
Unknown
33.1%
0.8%
c4ai-aya-expanse-32b
Unknown
32.8%
0.8%
llama-3-70b-instruct
Meta
32.5%
0.8%
qwen-image-prompt-extend
Qwen
32.2%
0.8%
Granite 4.0 H Small
Unknown
16.6%
15.6%
0.8%
Kling 1.5 Pro
Unknown
31.9%
0.8%
Vivago 2.0
Unknown
31.7%
0.8%
kling-v2.1-standard
Unknown
31.6%
0.8%
olmo-2-0325-32b-instruct
Unknown
31.5%
0.8%
reka-flash-20240904
Unknown
31.5%
0.8%
Qwen2-1.5B-Instruct
Qwen
7%
3.1%
2.3%
5.6%
5.5%
7.9%
0.8%
Wan 2.1 14B Open Weights
Unknown
25%
6.3%
0.8%
Yi-6B-Chat
Unknown
3.9%
5.4%
4.6%
8.8%
0%
7.9%
0.8%
T2V-01-Director
Unknown
30.6%
0.8%
command-r-plus
Cohere
30.3%
0.8%
hunyuan-standard-256k
Tencent
29.7%
0.7%
Pika 2.0
Unknown
29.2%
0.7%
Claude 3.5 Sonnet (June)
Anthropic
29.2%
0.7%
llama-3.1-tulu-3-8b
Meta
29%
0.7%
kandinsky-5.0-t2v-pro
Unknown
28.9%
0.7%
wan-v2.2-a14b
Unknown
15.8%
13.2%
0.7%
imagen-3.0-generate-002
Unknown
28.8%
0.7%
deepseek-coder-v2
DeepSeek
28.4%
0.7%
ministral-8b-2410
Mistral
28.4%
0.7%
qwen1.5-4b-chat
Qwen
4.1%
6.3%
3.8%
5.4%
2.4%
6.3%
0%
0.7%
p-image-edit
Unknown
28.3%
0.7%
llama-3.1-8b-instruct
Meta
27.4%
0.7%
qwen-image
Qwen
27.1%
0.7%
c4ai-aya-expanse-8b
Unknown
26.8%
0.7%
Kling 1.6 Standard
Unknown
26.4%
0.7%
T2V-01
Unknown
26.4%
0.7%
ltx-2-19b
Unknown
18.4%
7.9%
0.7%
seedance-v1-lite
Unknown
10.5%
15.8%
0.7%
reve-v1.1-fast
Unknown
26.1%
0.7%
yi-1.5-34b-chat
Unknown
25.9%
0.6%
reka-flash-21b-20240226-online
Unknown
25.6%
0.6%
ideogram-v3-quality
Unknown
25.4%
0.6%
ppl-sonar-reasoning-pro-high
Unknown
25%
0.6%
qwen-vl-max-1119
Qwen
25%
0.6%
qwen2-vl-72b
Qwen
25%
0.6%
llama-3-8b-instruct
Meta
24.9%
0.6%
reka-flash-21b-20240226
Unknown
24.3%
0.6%
Llama-2-7b-chat-hf
Meta
1.6%
6.2%
6.9%
6.4%
3.1%
0%
0.6%
command-r
Cohere
24%
0.6%
reve-edit-fast
Unknown
23.9%
0.6%
Nova Pro
Unknown
23.2%
0.6%
step-1v-32k
StepFun
23.1%
0.6%
internlm2_5-20b-chat
Unknown
23%
0.6%
gemma-2-2b-it
Google
22.7%
0.6%
granite-3.1-8b-instruct
Unknown
22.4%
0.6%
zephyr-orpo-141b-A35b-v0.1
Unknown
22.1%
0.6%
p-image
Unknown
22%
0.6%
photon
Unknown
22%
0.6%
qwen1.5-32b-chat
Qwen
21.5%
0.5%
molmo-72b-0924
Unknown
20.2%
0.5%
Runway Gen 3 Alpha
Unknown
15.3%
4.8%
0.5%
qwen1.5-14b-chat
Qwen
19.9%
0.5%
yi-34b-chat
Unknown
19.9%
0.5%
granite-3.1-2b-instruct
Unknown
19.6%
0.5%
Solar Mini
Unknown
19.6%
0.5%
tulu-2-dpo-70b
Unknown
19.2%
0.5%
hunyuan-standard-vision-2024-12-31
Tencent
19.2%
0.5%
wizardlm-70b
Unknown
18.9%
0.5%
ray2
Unknown
5.3%
5.3%
8.3%
0.5%
dbrx-instruct-preview
Unknown
18.6%
0.5%
llama-2-70b-chat
Meta
18.3%
0.5%
llama-3.2-vision-90b-instruct
Meta
18.3%
0.5%
nous-hermes-2-mixtral-8x7b-dpo
Unknown
18%
0.4%
I2V-01-Director
Unknown
17.5%
0.4%
wan2.5-i2i-preview
Unknown
17.4%
0.4%
llama-3.2-3b-instruct
Meta
17.4%
0.4%
qwen2-vl-7b-instruct
Qwen
17.3%
0.4%
starling-lm-7b-alpha
Unknown
17%
0.4%
recraft-v3
Unknown
16.9%
0.4%
openchat-3.5-0106
Unknown
16.7%
0.4%
Hunyuan Video (Fal) Open Weights
Tencent
16.7%
0%
0.4%
Mochi 1 Open Weights
Unknown
16.7%
0.4%
Pika 2.2
Unknown
6.9%
9.5%
0.4%
flux-1-kontext-dev
Unknown
3.4%
13%
0.4%
pixtral-12b-2409
Unknown
16.3%
0.4%
deepseek-llm-67b-chat
DeepSeek
16.1%
0.4%
vicuna-33b
Unknown
16.1%
0.4%
LTX Video v0.9.7 13B Open Weights
Unknown
15.9%
0.4%
snowflake-arctic-instruct
Unknown
15.8%
0.4%
llama2-70b-steerlm-chat
Unknown
15.5%
0.4%
internvl2-26b
Unknown
15.4%
0.4%
flux-1.1-pro
Unknown
15.3%
0.4%
openchat-3.5
Unknown
15.1%
0.4%
granite-3.0-8b-instruct
Unknown
14.8%
0.4%
Wan 2.2 5B Open Weights
Unknown
11.1%
3.2%
0.4%
Ray 1
Unknown
13.9%
0.3%
llama-2-13b-chat
Meta
13.2%
0.3%
kandinsky-5.0-t2v-lite
Unknown
13.2%
0.3%
solar-10.7b-instruct-v1.0
Unknown
12.6%
0.3%
Krea Realtime Open Weights
Unknown
12.5%
0.3%
dolphin-2.2.1-mistral-7b
Unknown
12.3%
0.3%
Granite 3.3 8B
Unknown
4.1%
8%
0.3%
ideogram-v2
Unknown
11.9%
0.3%
lucid-origin
Unknown
11.9%
0.3%
Gemma 3 27B
Google
11.8%
0.3%
Llama 2 Chat 7B
Meta
11.8%
0.3%
Mistral Large (Feb)
Mistral
0%
0%
11.8%
0.3%
Reka Flash 3
Unknown
11.8%
0.3%
granite-3.0-2b-instruct
Unknown
11.7%
0.3%
phi-3-mini-4k-instruct-june-2024
Unknown
11.7%
0.3%
yi-vision
Unknown
11.5%
0.3%
wizardlm-13b
Unknown
11.4%
0.3%
perplexity-sonar-reasoning
Unknown
4.7%
6.3%
0.3%
Qwen2-0.5B-Instruct
Qwen
2.3%
2.3%
0.8%
1.6%
3.9%
0%
0.3%
seededit-3.0
Unknown
10.9%
0.3%
ppl-sonar-pro-high
Unknown
10.7%
0.3%
mpt-30b-chat
Unknown
10.4%
0.3%
glm-image
Zhipu
10.2%
0.3%
codellama-34b-instruct
Unknown
10.1%
0.3%
BAAI bge-large-en-v1.5
BAAI
10%
0.3%
Kling 1.0
Unknown
9.7%
0.2%
c4ai-aya-vision-32b
Unknown
9.6%
0.2%
Gemma 3 12B
Google
9.6%
0.2%
Llama 3 70B
Meta
9.6%
0.2%
vicuna-13b
Unknown
9.5%
0.2%
codellama-70b-instruct
Unknown
9.1%
0.2%
gemma-7b-it
Google
8.8%
0.2%
molmo-7b-d-0924
Unknown
8.7%
0.2%
llama-3.2-1b-instruct
Meta
8.5%
0.2%
falcon-180b-chat
Unknown
8.2%
0.2%
Runway Gen 3 Alpha Turbo
Unknown
7.9%
0.2%
llama-3.2-vision-11b-instruct
Meta
7.7%
0.2%
guanaco-33b
Unknown
7.6%
0.2%
llama-2-7b-chat
Meta
7.6%
0.2%
qwen-14b-chat
Qwen
7.3%
0.2%
flux-1-dev-fp8
Unknown
6.8%
0.2%
nvila-internal-15b-v1
Unknown
6.7%
0.2%
smollm2-1.7b-instruct
Unknown
6.6%
0.2%
stripedhyena-nous-7b
Unknown
6.3%
0.2%
olmo-7b-instruct
Unknown
6%
0.1%
Apertus 70B Instruct
Unknown
5.9%
0.1%
LFM2 2.6B
Unknown
5.9%
0.1%
LFM2.5-1.2B-Instruct
Unknown
5.9%
0.1%
Mixtral 8x7B
Unknown
5.9%
0.1%
Olmo 3 7B
Unknown
5.9%
0.1%
Sarvam M
Unknown
5.9%
0.1%
llava-onevision-qwen2-72b-ov
Unknown
5.8%
0.1%
vicuna-7b
Unknown
5.7%
0.1%
Qwen1.5-1.8B-Chat
Qwen
0%
1.5%
3.1%
0.8%
0%
0%
0.1%
palm-2
Unknown
5.4%
0.1%
dall-e-3
Unknown
5.1%
0.1%
llava-v1.6-34b
Unknown
4.8%
0.1%
gemma-1.1-2b-it
Google
4.7%
0.1%
gemma-2b-it
Google
4.4%
0.1%
Haiper 2.0
Unknown
4.2%
0.1%
Pika 1.5
Unknown
4.2%
0.1%
LFM2 8B A1B
Unknown
4.1%
0.1%
koala-13b
Unknown
3.8%
0.1%
diffbot-small-xl
Unknown
3.6%
0.1%
chatglm3-6b
Unknown
3.5%
0.1%
gpt4all-13b-snoozy
Unknown
3.2%
0.1%
cogvlm2-llama3-chat-19b
Unknown
2.9%
0.1%
minicpm-v-2_6
Unknown
2.9%
0.1%
mpt-7b-chat
Unknown
2.8%
0.1%
Step-Video-T2V
StepFun
2.8%
0.1%
pika-v2.2
Unknown
2.6%
0%
0.1%
runway-gen4-turbo
Unknown
2.6%
0.1%
RWKV-4-Raven-14B
Unknown
2.5%
0.1%
Qwen1.5-0.5B-Chat
Qwen
0%
0.8%
1.5%
0%
0%
0%
0.1%
chatglm2-6b
Unknown
2.2%
0.1%
bagel
Unknown
0%
2.2%
0.1%
internvl2-4b
Unknown
1.9%
0%
alpaca-13b
Unknown
1.9%
0%
stable-diffusion-v35-large
Unknown
1.7%
0%
chatglm-6b
Unknown
1.6%
0%
CogVideoX-5B Open Weights
Unknown
1.4%
0%
oasst-pythia-12b
Unknown
1.3%
0%
phi-3.5-vision-instruct
Unknown
1%
0%
fastchat-t5-3b
Unknown
0.9%
0%
Apertus 8B Instruct
Unknown
0.7%
0%
Gemma 3 1B
Google
0.7%
0%
Gemma 3 4B
Google
0.7%
0%
LFM2 1.2B
Unknown
0.7%
0%
LFM2.5-VL-1.6B
Unknown
0.7%
0%
Llama 3 8B
Meta
0.7%
0%
stablelm-tuned-alpha-7b
Unknown
0.6%
0%
dolly-v2-12b
Unknown
0.3%
0%
Codestral
Mistral
0%
Codestral Embed
Mistral
0%
devstral-medium-2507
Mistral
0%
0%
0%
Gemma 3n E2B
Google
0%
0%
GPT Realtime 1.5
OpenAI
0%
GPT Realtime mini
OpenAI
0%
GPT-4o mini Transcribe
OpenAI
0%
GPT-4o mini TTS
OpenAI
0%
GPT-4o Transcribe
OpenAI
0%
HunyuanImage 3.0 Instruct (Fal)
Tencent
0%
0%
Leanstral
Mistral
0%
Llama 4 Behemoth
Meta
0%
llama-13b
Meta
0%
0%
Mistral Embed
Mistral
0%
Mistral Moderation 2
Mistral
0%
mochi-v1
Unknown
0%
0%
OCR 3
Mistral
0%
phi-3-vision-128k-instruct
Unknown
0%
0%
Pyramid Flow Open Weights
Unknown
0%
0%
Qwen3-Coder Plus
Qwen
0%
runway-gen4-aleph
Unknown
0%
0%
Seedream 4.0
Unknown
0%
0%
step1x-edit
Unknown
0%
0%
Tiny Aya Global
Unknown
0%
0%
Voxtral Mini Transcribe 2
Mistral
0%
Voxtral Mini Transcribe Realtime
Mistral
0%
Voxtral Small
Mistral
0%
Voxtral TTS
Mistral
0%
weakerstronger percentile inside group