~ same prompt, different LLM — every model gets the same brief in Blender, you judge the render ~
HomeLeaderboardGallerySandboxTasksVote!RunsAbout
Google / Gemini 3.8 Flash / thinking:medium has entered the arena NEW!
LeaderboardGalleryVoteRunsTasks
Task:Allabandoned-villagecabin-pondcamp-lanterncg-still-lifecorgi-beach-ballcurl-flowgoldfishlight-through-glassparam-towerplaster-bustpunched-cuberamen-samurairavensthwaite-terminalreaction-diffusion-relicred-ballrobotic-armshrine-vending-machinesmoke-cubespace-cowboy-duelspaceshipspectral-scannertext-assemblywave-gridworn-boot3 contestants | 24 votes | BT over 23 decisive
Mode:FullOne-shot3D-only
RankContestantRatingVotesRunsDone %Success %Avg $Avg callsAvg time
1OpenAI / GPT 5.6 Luna / thinking:maxcompare →1834231100100$0.0321427s
2OpenAI / GPT 5.6 Lunacompare →1166231100100$0.045106s
provOx Alpha / thinking:lowcompare →—0200$0.0018910s

Model comparison

each point is a model (Full, goldfish) — hover for exact values
OpenAIOther unrated
10661283150017171934$0.034$0.035$0.037$0.039$0.042Bradley–Terry ratingUNRATED — no decisive votes yetOpenAI / GPT 5.6 Luna / thinking:maxOpenAI / GPT 5.6 LunaMean cost per task (USD, log scale)

By dimension

separate Bradley–Terry fits from lens rounds + chip verdicts
Lighting
provOpenAI / GPT 5.6 Luna / thinking:max1595
provOpenAI / GPT 5.6 Luna1405
Composition
provOpenAI / GPT 5.6 Luna / thinking:max1595
provOpenAI / GPT 5.6 Luna1405
Modeling
provOpenAI / GPT 5.6 Luna / thinking:max1595
provOpenAI / GPT 5.6 Luna1405
Materials
provOpenAI / GPT 5.6 Luna / thinking:max1595
provOpenAI / GPT 5.6 Luna1405
Render
provOpenAI / GPT 5.6 Luna / thinking:max1595
provOpenAI / GPT 5.6 Luna1405
Adherence
provOpenAI / GPT 5.6 Luna / thinking:max1595
provOpenAI / GPT 5.6 Luna1405

Modes compared

each contestant's ratings across run modes (mixed ladder)
ContestantFullOne-shot3D-only
DeepSeek / Deepseek V4 Flash—1126—
DeepSeek / Deepseek V4 Flash 0731 / thinking:max———
Google / Gemini 3.5 Flash / thinking:medium—1114—
OpenAI / GPT 5.6 Luna13961447—
OpenAI / GPT 5.6 Luna / thinking:max2065——
OpenAI / GPT 5.6 Sol / thinking:max—1760—
OpenAI / GPT 6 Astra / thinking:medium—1715—
Anthropic / Opus 5 / thinking:medium—1644—
Ox Alpha / thinking:low———
Anthropic / Sonnet 5 / thinking:low—1233—
Vote A/BGalleryRunsbradley-terry fit over 23 decisive votes
★ Source on GitHub SOON · ♥ Support the bench
Votes cast on this bench: 001671 · 163 runs · 26 contestants
Best viewed with Netscape Navigator 4.07 at 800×600 · real runs, live ratings