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
330 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→
330 models ranked8901 official head-to-head verdictsPairwise tie-break active

Official standings

The rankings

226–250 of 330 results
#2261444 rating ±38
GPT-4o (2024-05-13) output

GPT-4o (2024-05-13)

OpenAI / May 13, 2024 / openrouter

Busey likeness1.2
Face coherence4.8
Aesthetics3.7
Prompt adherence3.4

1 judged output · judge spread σ 0.21 · W 16 / L 20 / D 50

2.6overall
#2271380 rating ±37
Qwen3.5-9B output

Qwen3.5-9B

Qwen / Mar 2, 2026 / openrouter

Busey likeness1.2
Face coherence4.3
Aesthetics3.6
Prompt adherence3.9

1 judged output · judge spread σ 0.29 · W 16 / L 30 / D 44

2.6overall
#2281351 rating ±39
Laguna M.1 output

Laguna M.1

Poolside / Apr 28, 2026 / openrouter

Busey likeness1.6
Face coherence3.9
Aesthetics3.1
Prompt adherence4.1

1 judged output · judge spread σ 0.36 · W 12 / L 37 / D 33

2.6overall
#2291322 rating ±37
Qwen3 30B A3B Instruct 2507 output

Qwen3 30B A3B Instruct 2507

Qwen / Jul 29, 2025 / openrouter

Busey likeness1.1
Face coherence4.2
Aesthetics3.5
Prompt adherence4.5

1 judged output · judge spread σ 0.68 · W 9 / L 38 / D 41

2.6overall
#2301572 rating ±35
GPT-3.5 Turbo (older v0613) output

GPT-3.5 Turbo (older v0613)

OpenAI / Jun 13, 2023 / openrouter

Busey likeness0.9
Face coherence4.4
Aesthetics3.9
Prompt adherence3.7

1 judged output · judge spread σ 0.39 · W 44 / L 8 / D 50

2.5overall
#2311567 rating ±35
Mistral Large 2407 output

Mistral Large 2407

Mistral / Jul 24, 2024 / openrouter

Busey likeness1.3
Face coherence3.6
Aesthetics3.3
Prompt adherence4.1

1 judged output · judge spread σ 0.21 · W 46 / L 10 / D 46

2.5overall
#2321479 rating ±35
Codestral 2508 output

Codestral 2508

Mistral / Aug 1, 2025 / openrouter

Busey likeness1.2
Face coherence4.2
Aesthetics3.1
Prompt adherence3.6

1 judged output · judge spread σ 0.38 · W 37 / L 22 / D 43

2.5overall
#2331418 rating ±35
GPT-4.1 Mini output

GPT-4.1 Mini

OpenAI / Apr 14, 2025 / openrouter

Busey likeness1.0
Face coherence4.5
Aesthetics3.8
Prompt adherence3.6

1 judged output · judge spread σ 0.34 · W 23 / L 25 / D 50

2.5overall
#2341391 rating ±35
GLM 4.5 Air output

GLM 4.5 Air

Z.ai / Jul 25, 2025 / openrouter

Busey likeness1.3
Face coherence3.9
Aesthetics3.2
Prompt adherence3.9

1 judged output · judge spread σ 0.46 · W 29 / L 37 / D 34

2.5overall
#2351255 rating ±36
GPT-4o (2024-08-06) output

GPT-4o (2024-08-06)

OpenAI / Aug 6, 2024 / openrouter

Busey likeness1.2
Face coherence4.6
Aesthetics3.3
Prompt adherence3.4

1 judged output · judge spread σ 0.31 · W 6 / L 45 / D 45

2.5overall
#2361214 rating ±35
Saba output

Saba

Mistral / Feb 17, 2025 / openrouter

Busey likeness1.1
Face coherence4.0
Aesthetics2.9
Prompt adherence4.6

1 judged output · judge spread σ 0.65 · W 9 / L 56 / D 35

2.5overall
#2371641 rating ±35
o3 Mini High output

o3 Mini High

OpenAI / Jan 31, 2025 / openrouter

Busey likeness1.0
Face coherence4.0
Aesthetics2.9
Prompt adherence4.4

1 judged output · judge spread σ 0.51 · W 60 / L 3 / D 37

2.4overall
#2381502 rating ±35
Qwen3 235B A22B Thinking 2507 output

Qwen3 235B A22B Thinking 2507

Qwen / Jul 25, 2025 / openrouter

Busey likeness0.9
Face coherence3.9
Aesthetics3.5
Prompt adherence4.0

1 judged output · judge spread σ 0.55 · W 46 / L 14 / D 42

2.4overall
#2391446 rating ±35
Mistral Medium 3.5 output

Mistral Medium 3.5

Mistral / Apr 30, 2026 / openrouter

Busey likeness1.3
Face coherence3.4
Aesthetics2.8
Prompt adherence4.4

1 judged output · judge spread σ 0.54 · W 28 / L 10 / D 64

2.4overall
#2401425 rating ±35
Qwen3 Coder 30B A3B Instruct output

Qwen3 Coder 30B A3B Instruct

Qwen / Jul 31, 2025 / openrouter

Busey likeness1.1
Face coherence4.1
Aesthetics3.0
Prompt adherence4.0

1 judged output · judge spread σ 0.62 · W 31 / L 23 / D 48

2.4overall
#2411412 rating ±35
GPT-4.1 Nano output

GPT-4.1 Nano

OpenAI / Apr 14, 2025 / openrouter

Busey likeness1.2
Face coherence4.2
Aesthetics2.6
Prompt adherence4.0

1 judged output · judge spread σ 0.36 · W 25 / L 22 / D 55

2.4overall
#2421384 rating ±34
Devstral 2 2512 output

Devstral 2 2512

Mistral / Dec 9, 2025 / openrouter

Busey likeness1.3
Face coherence3.7
Aesthetics2.8
Prompt adherence3.9

1 judged output · judge spread σ 0.52 · W 27 / L 23 / D 54

2.4overall
#2431270 rating ±35
Qwen3 Next 80B A3B Thinking output

Qwen3 Next 80B A3B Thinking

Qwen / Sep 11, 2025 / openrouter

Busey likeness1.3
Face coherence3.6
Aesthetics3.0
Prompt adherence4.0

1 judged output · judge spread σ 0.62 · W 16 / L 42 / D 44

2.4overall
#2441195 rating ±35
Ministral 3 14B 2512 output

Ministral 3 14B 2512

Mistral / Dec 2, 2025 / openrouter

Busey likeness1.3
Face coherence3.5
Aesthetics3.0
Prompt adherence4.1

1 judged output · judge spread σ 0.39 · W 7 / L 50 / D 43

2.4overall
#2451181 rating ±35
Nova Premier 1.0 output

Nova Premier 1.0

Amazon / Apr 30, 2025 / openrouter

Busey likeness1.2
Face coherence3.5
Aesthetics3.3
Prompt adherence3.7

1 judged output · judge spread σ 0.42 · W 5 / L 48 / D 49

2.4overall
#2461161 rating ±35
Sonar output

Sonar

Perplexity / Jan 21, 2025 / openrouter

Busey likeness1.0
Face coherence4.0
Aesthetics3.6
Prompt adherence3.6

1 judged output · judge spread σ 0.38 · W 11 / L 60 / D 31

2.4overall
#2471139 rating ±36
DeepSeek V3 0324 output

DeepSeek V3 0324

DeepSeek / Mar 24, 2025 / openrouter

Busey likeness1.0
Face coherence4.0
Aesthetics3.0
Prompt adherence3.9

1 judged output · judge spread σ 0.46 · W 4 / L 47 / D 41

2.4overall
#2481122 rating ±36
Step 3.5 Flash output

Step 3.5 Flash

StepFun / Jan 29, 2026 / openrouter

Busey likeness1.1
Face coherence3.5
Aesthetics3.5
Prompt adherence3.9

1 judged output · judge spread σ 0.57 · W 2 / L 48 / D 42

2.4overall
#2491077 rating ±35
Llama 4 Maverick output

Llama 4 Maverick

Meta / Apr 5, 2025 / openrouter

Busey likeness1.1
Face coherence3.4
Aesthetics4.0
Prompt adherence4.0

1 judged output · judge spread σ 0.65 · W 2 / L 68 / D 32

2.4overall
#250976 rating ±35
R1 output

R1

DeepSeek / Jan 20, 2025 / openrouter

Busey likeness1.2
Face coherence3.7
Aesthetics3.2
Prompt adherence3.6

1 judged output · judge spread σ 0.43 · W 3 / L 80 / D 19

2.4overall
← PreviousPage 10 of 14Next →
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