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

226–250 of 306 results
#2261030 rating ±45
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 1 / L 47 / D 12

2.4overall
#2271591 rating ±46
Perceptron Mk1 output

Perceptron Mk1

Perceptron / May 12, 2026 / openrouter

Busey likeness0.7
Face coherence4.9
Aesthetics2.9
Prompt adherence3.7

1 judged output · judge spread σ 0.33 · W 30 / L 1 / D 28

2.3overall
#2281503 rating ±46
Step 3.7 Flash output

Step 3.7 Flash

StepFun / May 28, 2026 / openrouter

Busey likeness0.8
Face coherence3.7
Aesthetics3.9
Prompt adherence3.8

1 judged output · judge spread σ 0.46 · W 22 / L 5 / D 32

2.3overall
#2291486 rating ±45
Ling-2.6-1T output

Ling-2.6-1T

inclusionAI / Apr 23, 2026 / openrouter

Busey likeness0.8
Face coherence3.3
Aesthetics4.6
Prompt adherence3.7

1 judged output · judge spread σ 0.45 · W 21 / L 4 / D 35

2.3overall
#2301468 rating ±46
Qwen3 Coder Flash output

Qwen3 Coder Flash

Qwen / Jul 31, 2025 / openrouter

Busey likeness0.9
Face coherence3.7
Aesthetics3.4
Prompt adherence4.2

1 judged output · judge spread σ 0.49 · W 22 / L 10 / D 27

2.3overall
#2311427 rating ±45
R1 0528 output

R1 0528

DeepSeek / May 28, 2025 / openrouter

Busey likeness1.4
Face coherence2.8
Aesthetics3.1
Prompt adherence3.9

1 judged output · judge spread σ 0.55 · W 17 / L 11 / D 32

2.3overall
#2321409 rating ±45
Qwen3 VL 8B Instruct output

Qwen3 VL 8B Instruct

Qwen / Oct 14, 2025 / openrouter

Busey likeness0.8
Face coherence3.9
Aesthetics3.6
Prompt adherence3.6

1 judged output · judge spread σ 0.88 · W 14 / L 10 / D 36

2.3overall
#2331299 rating ±46
GPT-4o-mini Search Preview output

GPT-4o-mini Search Preview

OpenAI / Mar 12, 2025 / openrouter

Busey likeness1.1
Face coherence3.7
Aesthetics2.9
Prompt adherence3.9

1 judged output · judge spread σ 0.40 · W 4 / L 17 / D 38

2.3overall
#2341202 rating ±46
Nemotron 3 Nano 30B A3B output

Nemotron 3 Nano 30B A3B

NVIDIA / Dec 14, 2025 / openrouter

Busey likeness0.9
Face coherence4.2
Aesthetics2.8
Prompt adherence3.7

1 judged output · judge spread σ 0.40 · W 5 / L 31 / D 23

2.3overall
#2351683 rating ±45
LongCat 2.0 output

LongCat 2.0

Meituan / Jul 20, 2026 / openrouter

Busey likeness0.9
Face coherence3.5
Aesthetics3.5
Prompt adherence3.6

1 judged output · judge spread σ 0.86 · W 40 / L 0 / D 20

2.2overall
#2361551 rating ±45
GLM 4.5V output

GLM 4.5V

Z.ai / Aug 11, 2025 / openrouter

Busey likeness1.1
Face coherence3.4
Aesthetics2.9
Prompt adherence3.7

1 judged output · judge spread σ 0.57 · W 30 / L 2 / D 28

2.2overall
#2371421 rating ±45
GLM 4.6V output

GLM 4.6V

Z.ai / Dec 8, 2025 / openrouter

Busey likeness1.2
Face coherence3.2
Aesthetics2.8
Prompt adherence3.6

1 judged output · judge spread σ 0.35 · W 21 / L 14 / D 25

2.2overall
#2381281 rating ±45
Nemotron 3 Nano Omni (free) output

Nemotron 3 Nano Omni (free)

NVIDIA / Apr 28, 2026 / openrouter

Busey likeness1.0
Face coherence4.2
Aesthetics2.5
Prompt adherence3.1

1 judged output · judge spread σ 0.45 · W 9 / L 16 / D 35

2.2overall
#2391267 rating ±45
Granite 4.2 8B output

Granite 4.2 8B

IBM / Aug 25, 2026 / openrouter

Busey likeness1.1
Face coherence3.9
Aesthetics2.8
Prompt adherence3.1

1 judged output · judge spread σ 0.59 · W 8 / L 22 / D 30

2.2overall
#2401230 rating ±45
Jamba Large 1.7 output

Jamba Large 1.7

AI21 / Aug 8, 2025 / openrouter

Busey likeness1.1
Face coherence3.2
Aesthetics3.4
Prompt adherence3.3

1 judged output · judge spread σ 0.45 · W 10 / L 23 / D 27

2.2overall
#2411098 rating ±45
MiniMax M2-her output

MiniMax M2-her

MiniMax / Jan 23, 2026 / openrouter

Busey likeness1.4
Face coherence2.5
Aesthetics2.7
Prompt adherence3.7

1 judged output · judge spread σ 0.41 · W 2 / L 33 / D 25

2.2overall
#2421699 rating ±45
Qwen3 235B A22B output

Qwen3 235B A22B

Qwen / Apr 28, 2025 / openrouter

Busey likeness0.9
Face coherence3.7
Aesthetics2.8
Prompt adherence3.3

1 judged output · judge spread σ 0.66 · W 47 / L 1 / D 12

2.1overall
#2431388 rating ±45
Command A output

Command A

Cohere / Mar 13, 2025 / openrouter

Busey likeness0.9
Face coherence3.4
Aesthetics2.8
Prompt adherence3.6

1 judged output · judge spread σ 0.49 · W 21 / L 12 / D 27

2.1overall
#2441349 rating ±45
Nemotron 3.5 Lightning output

Nemotron 3.5 Lightning

NVIDIA / Aug 11, 2026 / openrouter

Busey likeness0.9
Face coherence4.3
Aesthetics2.6
Prompt adherence3.0

1 judged output · judge spread σ 0.65 · W 20 / L 15 / D 25

2.1overall
#2451329 rating ±45
Hy-MT2-7B output

Hy-MT2-7B

Tencent / May 21, 2026 / openrouter

Busey likeness0.7
Face coherence3.6
Aesthetics3.3
Prompt adherence3.2

1 judged output · judge spread σ 0.78 · W 9 / L 9 / D 42

2.1overall
#2461297 rating ±45
Hy3 preview output

Hy3 preview

Tencent / Apr 22, 2026 / openrouter

Busey likeness0.8
Face coherence3.4
Aesthetics3.9
Prompt adherence3.1

1 judged output · judge spread σ 0.44 · W 11 / L 16 / D 33

2.1overall
#2471252 rating ±45
GPT-4o-mini output

GPT-4o-mini

OpenAI / Jul 18, 2024 / openrouter

Busey likeness0.5
Face coherence4.2
Aesthetics3.3
Prompt adherence3.0

1 judged output · judge spread σ 0.45 · W 11 / L 22 / D 27

2.1overall
#2481183 rating ±45
Llama 3.3 70B Instruct output

Llama 3.3 70B Instruct

Meta / Dec 6, 2024 / openrouter

Busey likeness0.9
Face coherence4.2
Aesthetics2.8
Prompt adherence3.1

1 judged output · judge spread σ 0.31 · W 0 / L 20 / D 40

2.1overall
#2491447 rating ±49
Qwen3 VL 235B A22B Instruct output

Qwen3 VL 235B A22B Instruct

Qwen / Sep 23, 2025 / openrouter

Busey likeness0.9
Face coherence2.9
Aesthetics2.9
Prompt adherence3.4

1 judged output · judge spread σ 0.27 · W 24 / L 4 / D 24

2.0overall
#2501431 rating ±48
GPT-3.5 Turbo 16k output

GPT-3.5 Turbo 16k

OpenAI / Jun 13, 2023 / openrouter

Busey likeness0.7
Face coherence3.0
Aesthetics3.8
Prompt adherence3.1

1 judged output · judge spread σ 0.54 · W 22 / L 6 / D 26

2.0overall
← PreviousPage 10 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