Skip to main content
BuseyBenchcreative model benchmarkSVG models tested
SVG testImage testLeaderboardTimelineMethodology
Browse
SVG testImage testLeaderboardTimelineMethodology
BuseyBenchSame prompt. Same rules. Maximum facial voltage.
Score trendsTimelineMethodologyAboutPrivacyGovernanceRSSAdmin
306 SVG models tested
BuseyBench SVG TestImage Generator Portrait Test

Leaderboard

Judged. Ranked. Busey'd.

One output per model, independent judge scores, and close-score pairwise tie-breaks. The best Gary Busey portraits rise to the top.

Explore the dataDrawing ability over time→
306 models ranked7103 official head-to-head verdictsPairwise tie-break active

Official standings

The rankings

151–175 of 306 results
#1511576 rating ±45
Seed-2.0-Code output

Seed-2.0-Code

ByteDance Seed / Feb 14, 2026 / openrouter

Busey likeness1.6
Face coherence5.1
Aesthetics4.2
Prompt adherence5.4

1 judged output · judge spread σ 0.43 · W 14 / L 14 / D 32

3.3overall
#1521575 rating ±46
GPT-5 Codex output

GPT-5 Codex

OpenAI / Sep 15, 2025 / openrouter

Busey likeness2.0
Face coherence3.9
Aesthetics4.6
Prompt adherence5.6

1 judged output · judge spread σ 0.50 · W 14 / L 13 / D 32

3.3overall
#1531564 rating ±46
Gemini 2.5 Flash output

Gemini 2.5 Flash

Google / Jun 17, 2025 / openrouter

Busey likeness1.7
Face coherence5.1
Aesthetics4.6
Prompt adherence5.0

1 judged output · judge spread σ 0.79 · W 15 / L 17 / D 27

3.3overall
#1541427 rating ±46
Sonar Pro Search output

Sonar Pro Search

Perplexity / Oct 30, 2025 / openrouter

Busey likeness1.2
Face coherence5.4
Aesthetics5.2
Prompt adherence5.2

1 judged output · judge spread σ 0.30 · W 4 / L 25 / D 30

3.3overall
#1551667 rating ±46
Mistral Large 3 2512 output

Mistral Large 3 2512

Mistral / Dec 1, 2025 / openrouter

Busey likeness1.7
Face coherence4.6
Aesthetics4.1
Prompt adherence5.6

1 judged output · judge spread σ 0.41 · W 25 / L 9 / D 25

3.2overall
#1561520 rating ±45
DeepSeek V3.2 output

DeepSeek V3.2

DeepSeek / Dec 1, 2025 / openrouter

Busey likeness1.6
Face coherence5.0
Aesthetics4.4
Prompt adherence5.0

1 judged output · judge spread σ 0.67 · W 13 / L 20 / D 27

3.2overall
#1571489 rating ±45
Qwen Plus 0728 output

Qwen Plus 0728

Qwen / Jul 28, 2025 / openrouter

Busey likeness1.6
Face coherence5.5
Aesthetics3.9
Prompt adherence4.6

1 judged output · judge spread σ 0.51 · W 6 / L 18 / D 36

3.2overall
#1581346 rating ±46
Qwen3 Coder Next output

Qwen3 Coder Next

Qwen / Feb 4, 2026 / openrouter

Busey likeness1.6
Face coherence5.4
Aesthetics4.1
Prompt adherence4.7

1 judged output · judge spread σ 0.55 · W 4 / L 35 / D 20

3.2overall
#1591797 rating ±45
Gemini 2.5 Flash Lite Preview 09-2025 output

Gemini 2.5 Flash Lite Preview 09-2025

Google / Sep 25, 2025 / openrouter

Busey likeness1.4
Face coherence5.0
Aesthetics4.1
Prompt adherence5.3

1 judged output · judge spread σ 0.58 · W 38 / L 4 / D 18

3.1overall
#1601795 rating ±45
GPT-5 Mini output

GPT-5 Mini

OpenAI / Aug 7, 2025 / openrouter

Busey likeness1.6
Face coherence4.9
Aesthetics4.0
Prompt adherence4.9

1 judged output · judge spread σ 0.72 · W 37 / L 3 / D 21

3.1overall
#1611673 rating ±45
Gemini 3.1 Flash Lite output

Gemini 3.1 Flash Lite

Google / May 7, 2026 / openrouter

Busey likeness1.5
Face coherence5.0
Aesthetics3.8
Prompt adherence5.7

1 judged output · judge spread σ 0.54 · W 26 / L 10 / D 24

3.1overall
#1621624 rating ±45
Muse Glimmer 30B output

Muse Glimmer 30B

Meta / Aug 9, 2026 / openrouter

Busey likeness1.6
Face coherence4.7
Aesthetics4.5
Prompt adherence4.6

1 judged output · judge spread σ 0.67 · W 22 / L 11 / D 27

3.1overall
#1631570 rating ±45
DeepSeek V3.1 output

DeepSeek V3.1

DeepSeek / Aug 21, 2025 / openrouter

Busey likeness1.3
Face coherence4.5
Aesthetics5.1
Prompt adherence5.4

1 judged output · judge spread σ 0.42 · W 14 / L 13 / D 33

3.1overall
#1641565 rating ±45
Mercury 2.5 output

Mercury 2.5

Inception / Aug 31, 2026 / openrouter

Busey likeness2.0
Face coherence4.4
Aesthetics4.1
Prompt adherence4.2

1 judged output · judge spread σ 0.69 · W 20 / L 20 / D 20

3.1overall
#1651563 rating ±45
Cogito v2.1 671B output

Cogito v2.1 671B

Deep Cogito / Nov 13, 2025 / openrouter

Busey likeness1.3
Face coherence5.1
Aesthetics4.0
Prompt adherence5.4

1 judged output · judge spread σ 0.69 · W 18 / L 17 / D 25

3.1overall
#1661556 rating ±45
gpt-oss-120b output

gpt-oss-120b

OpenAI / Aug 5, 2025 / openrouter

Busey likeness1.3
Face coherence5.7
Aesthetics3.9
Prompt adherence4.9

1 judged output · judge spread σ 0.47 · W 17 / L 17 / D 26

3.1overall
#1671534 rating ±45
GLM 4.5 output

GLM 4.5

Z.ai / Jul 25, 2025 / openrouter

Busey likeness1.6
Face coherence5.1
Aesthetics3.6
Prompt adherence4.9

1 judged output · judge spread σ 0.68 · W 15 / L 18 / D 28

3.1overall
#1681494 rating ±45
Solar Pro 4 output

Solar Pro 4

Upstage / Aug 10, 2026 / openrouter

Busey likeness1.4
Face coherence4.8
Aesthetics4.6
Prompt adherence5.3

1 judged output · judge spread σ 0.44 · W 12 / L 24 / D 24

3.1overall
#1691714 rating ±46
Qwen3 Max Thinking output

Qwen3 Max Thinking

Qwen / Jan 25, 2026 / openrouter

Busey likeness1.3
Face coherence4.9
Aesthetics4.5
Prompt adherence4.4

1 judged output · judge spread σ 0.49 · W 28 / L 2 / D 29

3.0overall
#1701689 rating ±45
o4 Mini output

o4 Mini

OpenAI / Apr 16, 2025 / openrouter

Busey likeness1.3
Face coherence5.1
Aesthetics4.0
Prompt adherence5.3

1 judged output · judge spread σ 0.62 · W 25 / L 5 / D 30

3.0overall
#1711520 rating ±46
Qwen3 Max output

Qwen3 Max

Qwen / Sep 23, 2025 / openrouter

Busey likeness1.3
Face coherence5.6
Aesthetics4.0
Prompt adherence4.0

1 judged output · judge spread σ 0.46 · W 17 / L 20 / D 22

3.0overall
#1721477 rating ±46
Qwen-Plus output

Qwen-Plus

Qwen / May 1, 2024 / openrouter

Busey likeness1.3
Face coherence4.9
Aesthetics3.7
Prompt adherence5.6

1 judged output · judge spread σ 0.61 · W 10 / L 21 / D 28

3.0overall
#1731418 rating ±46
R1 Distill Llama 70B output

R1 Distill Llama 70B

DeepSeek / Jan 23, 2025 / openrouter

Busey likeness1.6
Face coherence4.1
Aesthetics3.6
Prompt adherence5.4

1 judged output · judge spread σ 0.83 · W 10 / L 28 / D 21

3.0overall
#1741244 rating ±46
GPT-5.3-Codex-Spark output

GPT-5.3-Codex-Spark

OpenAI / Feb 12, 2026 / manual upload

Busey likeness1.6
Face coherence4.5
Aesthetics4.7
Prompt adherence4.1

1 judged output · judge spread σ 0.67 · W 2 / L 43 / D 14

3.0overall
#1751798 rating ±46
Mistral Large output

Mistral Large

Mistral / Feb 26, 2024 / openrouter

Busey likeness1.5
Face coherence4.4
Aesthetics3.8
Prompt adherence4.6

1 judged output · judge spread σ 0.68 · W 40 / L 5 / D 14

2.9overall
← PreviousPage 7 of 13Next →
Ranking statusDisplayed score + pairwise tie-break

Overall = likeness 50% + face 20% + aesthetics 15% + prompt 15%. Each model contributes one canonical judged output. The visible overall score, rounded to one decimal, always sets the score band. Pairwise results only order models tied within that displayed band, so a lower displayed score can never leapfrog a higher one.

Read the full methodology →View score trends →
1canonical output judged per model
Closest-scorepairwise matchmaking strategy
busey-v2current scoring rubric