~ 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!
ARCHIVEDRetired from the leaderboard and voting — smoke-test scene — capability sanity check, not part of the benchmark · 2026-07-20. Preserved here for reference only.
Render Window — hero
hero render
camera + framing set up by the AI contestant
3D Viewport — red-ball User Persp
LOADING 3D VIEW
$0.07Cost
61sDuration
7Tool calls
8Turns
Blender 5.1.0MCP 1.22026-07-15
Tokens: 78074/2368Vision: Votes W-L-T: 0-0-0reasoning: effort lowDONE ✓
B-BENCH — TRANSCRIPT Anthropic / Sonnet 5 / thinking:low · r1 · 2026-07-15
RUN LOG — red-ball · Anthropic / Sonnet 5 / thinking:low · r1 · ok
SYSTEM PROMPTthe control — every model on this task got these instructions
PROMPTblender-bench v6HARNESSraw-v1
SHOW FULL PROMPT · 1104 chars
You are participating in Blender-Bench, a public benchmark where AI models complete the same Blender task.

Your result will be compared head-to-head with another model. Anyone on the internet can vote, based only on the published visual outputs they can inspect: the interactive 3D scene and rendered presentation.

Complete the request as well as you can. The benchmark does not prescribe a workflow or visual style; use whichever Blender techniques you consider appropriate. Make the important requirements evident in the published result, save a complete scene, and present the work clearly in every requested output.

## Harness Instructions

You are operating Blender through the provided tools. The scene starts empty. Build what the user's prompt asks for in the current scene. Work step by step and verify your own work. Before you finish, set up a camera and lighting to frame a single, well-composed presentation image of your result — position and aim the camera, and add or adjust lights so the subject reads clearly. When you are finished, reply with the single word DONE and no tool calls.
--TASK red-ball: ## Task Complete the following Blender task: <task> A red sphere of 1 meter diameter resting on a white ground plane — touching it exactly, neither floating nor sunk. Smooth shading, a plain red material on the ball, plain white on the floor. The floor is the sphere, the color, and the contact. The ceiling is presentation: a soft believable shadow under the ball, a hint of glossiness that catc…
07:19>>execute_blender_code({"code":"\nimport bpy\n\n# Clear scene\nbpy.ops.object.select_all(action='SELECT')\nbpy.ops.object.delete(use_global=False)\n\n# Ground plane\nbpy.ops.mesh.primitive_plane_add(size=10, location=(0,0,0))\nplane = bpy.context.active_object\nplane.name = \"Ground\"\n\nmat_white = bpy.data.materials.new(name=\"WhiteMat\")\nmat_white.use_nodes = True\nbsdf = mat_white.node_tree.nodes[\"Principled BSDF\…)
07:19<<Code executed successfully: done
07:28ai>Now let's add camera and lighting for a good presentation shot.
07:28>>execute_blender_code({"code":"\nimport bpy, math\n\n# Camera\nbpy.ops.object.camera_add(location=(2.8, -2.8, 1.6))\ncam = bpy.context.active_object\ncam.name = \"Camera\"\ndirection = (0,0,0.5)\n# point camera at ball\nimport mathutils\nloc = mathutils.Vector(cam.location)\ntarget = mathutils.Vector((0,0,0.4))\ndirection_vec = target - loc\nrot_quat = direction_vec.to_track_quat('-Z','Y')\ncam.rotation_euler = rot_qua…)
07:28<<Code executed successfully: camera/lights set
07:31>>get_viewport_screenshot({"user_prompt":"red sphere on white plane presentation"})
07:32<<
07:32imgimage delivered — get_viewport_screenshot
07:37ai>This is the viewport (solid shading, no materials rendered). Let me switch viewport shading to rendered to check materials/lighting look correct.
07:37>>execute_blender_code({"code":"\nimport bpy\nfor area in bpy.context.screen.areas:\n if area.type == 'VIEW_3D':\n for space in area.spaces:\n if space.type == 'VIEW_3D':\n space.shading.type = 'RENDERED'\n space.region_3d.view_perspective = 'CAMERA'\nbpy.context.scene.camera = bpy.data.objects['Camera']\nprint(\"set\")\n"})
07:37<<Code executed successfully: set
07:41>>get_viewport_screenshot({"user_prompt":"red sphere on white plane rendered view from camera"})
07:44<<
07:44imgimage delivered — get_viewport_screenshot
07:51ai>This looks good. The sphere shows nice glossy highlights, sits properly on the plane, red material reads well, and the white floor is clean. Contact shadow looks a bit weak though — let me check it's actually visible under the ball.
07:51>>execute_blender_code({"code":"\nimport bpy\n# Slightly reduce ambient world light influence and add contact shadow enhancement\n# Check world background isn't washing out the shadow\nworld = bpy.context.scene.world\nbg = world.node_tree.nodes[\"Background\"]\nbg.inputs[1].default_value = 0.3\n\n# Increase floor roughness slightly reduced for cleaner look, keep white\nmat_white = bpy.data.materials[\"WhiteMat\"]\nmat_w…)
07:52<<Code executed successfully: adjusted
07:56>>get_viewport_screenshot({"user_prompt":"final check of red sphere scene with shadow"})
07:59<<
07:59imgimage delivered — get_viewport_screenshot
08:04ai>This looks great — glossy red sphere resting exactly on the white plane with a soft contact shadow visible beneath, clean framing, and good specular highlights. The scene meets all requirements. DONE
08:04--status: ok · DONE declared
>
FULL CALL INSPECTOR
OPEN RAW JSONL ↗
0 SHOWN
LOADING COMPLETE TRANSCRIPT…
Votes cast on this bench: 000482 · 157 runs · 25 contestants
Best viewed with Netscape Navigator 4.07 at 800×600 · real runs, live ratings