~ same prompt, different LLM — every model gets the same brief in Blender, you judge the render ~
Google / Gemini 3.8 Flash / thinking:medium has entered the arena NEW!
LeaderboardGalleryVoteRunsTasks
RankContestantRatingVotesRunsDone %Success %Avg $Avg callsAvg time
1OpenAI / GPT 6 Astra / thinking:mediumcompare →18492510100$0.33—195s
2OpenAI / GPT 5.6 Sol / thinking:maxcompare →1672211100100$1.50—842s
3OpenAI / GPT 5.6 Lunacompare →16672210100$0.02—66s
4DeepSeek / Deepseek V4 Flashcompare →1450231100100$0.01—449s
5Google / Gemini 3.5 Flash / thinking:mediumcompare →13842310100$0.22—170s
6Google / Gemma 4 31b Itcompare →11712410100$0.00—152s
provAnthropic / Sonnet 5 / thinking:lowcompare →1307210100$0.04—82s
provGoogle / Gemini 3.6 Flash / thinking:mediumcompare →—010100$0.19—133s

Model comparison

each point is a model (One-shot, camp-lantern) — hover for exact values
AnthropicDeepSeekGoogleOpenAI unrated
10691290151017301951$0.000$0.004$0.034$0.308$2.75Bradley–Terry ratingUNRATED — no decisive votes yetOpenAI / GPT 6 Astra / thinking:mediumOpenAI / GPT 5.6 Sol / thinking:maxOpenAI / GPT 5.6 LunaDeepSeek / Deepseek V4 FlashGoogle / Gemini 3.5 Flash / thinking:mediumGoogle / Gemma 4 31b ItAnthropic / Sonnet 5 / thinking:lowMean cost per task (USD, log scale)

By dimension

lens rounds and chip verdicts feed these — none cast yet
Lighting
no data yet
Composition
no data yet
Modeling
no data yet
Materials
no data yet
Render
no data yet
Adherence
no data yet

Modes compared

each contestant's ratings across run modes (mixed ladder)
ContestantFullOne-shot3D-only
DeepSeek / Deepseek V4 Flash14471516—
DeepSeek / Deepseek V4 Pro1600——
Google / Gemini 3.1 Flash Lite1185——
Google / Gemini 3.5 Flash / thinking:medium16881446—
Google / Gemini 3.6 Flash / thinking:medium———
Google / Gemma 4 31b It12421252—
Z.AI / GLM 5.21508——
OpenAI / GPT 5.6 Luna15451729—
OpenAI / GPT 5.6 Luna / thinking:max———
OpenAI / GPT 5.6 Sol / thinking:max—1734—
OpenAI / GPT 6 Astra / thinking:medium—1913—
Anthropic / Haiku 4.5 / thinking:xhigh1334——
Anthropic / Sonnet 5 / thinking:low—1363—
Vote A/BGalleryRunsbradley-terry fit over 70 decisive votes
Votes cast on this bench: 001671 · 163 runs · 26 contestants
Best viewed with Netscape Navigator 4.07 at 800×600 · real runs, live ratings