~ same prompt, different LLM — every model gets the same brief in Blender, you judge the render ~
OpenAI / GPT 6 Astra / thinking:max has entered the arena NEW!
LeaderboardGalleryVoteRunsTasks
RankContestantRatingVotesRunsDone %Success %Avg $Avg callsAvg time
1OpenAI / GPT 5.6 Sol / thinking:maxcompare →175761100100$1.39943s
2Anthropic / Opus 5 / thinking:mediumcompare →170791100100$0.91490s
3OpenAI / GPT 5.6 Lunacompare →1545810100$0.0376s
4Anthropic / Sonnet 5 / thinking:lowcompare →1424810100$0.0797s
5DeepSeek / Deepseek V4 Flashcompare →1320810100$0.00129s
provOpenAI / GPT 6 Astra / thinking:mediumcompare →1513210100$0.51198s
provGoogle / Gemini 3.5 Flash / thinking:mediumcompare →1233110100$0.23150s
provDeepSeek / Deepseek V4 Flash 0731 / thinking:maxcompare →010100$0.01492s

Model comparison

each point is a model (One-shot, goldfish) — hover for exact values
AnthropicDeepSeekGoogleOpenAI unrated
11541325149516651836$0.001$0.008$0.054$0.356$2.34Bradley–Terry ratingUNRATED — no decisive votes yetOpenAI / GPT 5.6 Sol / thinking:maxAnthropic / Opus 5 / thinking:mediumOpenAI / GPT 5.6 LunaAnthropic / Sonnet 5 / thinking:lowDeepSeek / Deepseek V4 FlashOpenAI / GPT 6 Astra / thinking:mediumGoogle / Gemini 3.5 Flash / thinking:mediumMean 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 Flash1275
DeepSeek / Deepseek V4 Flash 0731 / thinking:max
Google / Gemini 3.5 Flash / thinking:medium1207
OpenAI / GPT 5.6 Luna14141499
OpenAI / GPT 5.6 Luna / thinking:max1885
OpenAI / GPT 5.6 Sol / thinking:max1709
OpenAI / GPT 6 Astra / thinking:medium1468
Anthropic / Opus 5 / thinking:medium1662
Ox Alpha / thinking:low
Anthropic / Sonnet 5 / thinking:low1382
Vote A/BGalleryRunsbradley-terry fit over 21 decisive votes
Votes cast on this bench: 000504 · 157 runs · 25 contestants
Best viewed with Netscape Navigator 4.07 at 800×600 · real runs, live ratings