~ 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 5.6 Luna / thinking:maxcompare →1790251100100$0.0215299s
2Z.AI / GLM 5.2compare →1653181100100$0.1829703s
3Google / Gemini 3.5 Flash / thinking:mediumcompare →1649201100100$2.0831562s
4OpenAI / GPT 5.6 Lunacompare →1621211100100$0.055103s
5DeepSeek / Deepseek V4 Procompare →142091100100$0.1540790s
6Anthropic / Haiku 4.5 / thinking:xhighcompare →1393121100100$0.2034352s
7Google / Gemini 3.1 Flash Litecompare →1351141100100$0.021291s
8Google / Gemma 4 31b Itcompare →1325131100100$0.049148s
9DeepSeek / Deepseek V4 Flashcompare →1296141100100$0.0329521s

Model comparison

each point is a model (Full, abandoned-village) — hover for exact values
AnthropicDeepSeekGoogleOpenAIZ.AI unrated
12221382154317041864$0.010$0.043$0.179$0.742$3.07Bradley–Terry ratingMean 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
Anthropic / Claude Fable 5.1 / thinking:medium—1590—
DeepSeek / Deepseek V4 Flash13501258—
DeepSeek / Deepseek V4 Flash 0731 / thinking:max—1164—
DeepSeek / Deepseek V4 Pro1474——
Google / Gemini 3.1 Flash Lite1408——
Google / Gemini 3.5 Flash / thinking:medium17001187—
Google / Gemini 3.6 Flash / thinking:medium—1451—
Google / Gemini 3.8 Flash / thinking:medium———
Google / Gemma 4 31b It13781475—
Z.AI / GLM 5.21706——
OpenAI / GPT 5.6 Luna16741249—
OpenAI / GPT 5.6 Luna / thinking:max1843——
OpenAI / GPT 5.6 Sol / thinking:max—1538—
OpenAI / GPT 6 Astra / thinking:medium—1874—
Anthropic / Haiku 4.5 / thinking:xhigh1446——
Anthropic / Opus 5 / thinking:medium—1792—
Anthropic / Sonnet 5 / thinking:low—1444—
Vote A/BGalleryRunsbradley-terry fit over 73 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