~ 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!
www.blender-bench.org   Ru:157
//runs/
Scene
robotic-arm__kimi-k3-think-max__r1__20260717-185914 renderAI cam
Desk-mounted robotic arm · r1 · 2026-07-17
Moonshot AI / Kimi K3 / thinking:maxDONE ✓
cg-still-life__kimi-k3-think-max__r1__20260717-174738 renderAI cam
Still life with the ancestors · r1 · 2026-07-17
Moonshot AI / Kimi K3 / thinking:maxMODEL_ERROR
cabin-pond__gpt-5.6-luna-think-max__r1__20260716-212158 renderAI cam
Wooden cabin by a pond · r1 · 2026-07-16
OpenAI / GPT 5.6 Luna / thinking:maxDONE ✓
cabin-pond__kimi-k3-think-max__r1__20260716-203205 renderauto
Wooden cabin by a pond · r1 · 2026-07-16
Moonshot AI / Kimi K3 / thinking:maxMODEL_ERROR
Compare tray — 1 of 2
A: light-through-glass · Google / Gemma 4 31b It · r1 CLEARCOMPARE →
Votes cast on this bench: 000535 · 157 runs · 25 contestants
Best viewed with Netscape Navigator 4.07 at 800×600 · real runs, live ratings