Blender-Bench — leaderboard

BLENDER-BENCH

~ which robot blends best? LLMs battle inside Blender, you pick the winner ~

Anthropic / Sonnet 5 / thinking:low has entered the arena NEW!

FileEditRenderHelpLeaderboardGallery Vote Runs Tasks

Task:All cabin-pondSighted ✓5 contestants | 0 votes | BT pending

Rank	Contestant	Rating	Runs	Done %	Avg $	Avg calls	Avg time
1	Anthropic / Sonnet 5 / thinking:lowcompare →	—	1	100	$0.34	38	294s
2	Z.AI / GLM 5.2compare →	—	1	100	$0.09	16	303s
3	OpenAI / GPT 5.6 Terracompare →	—	1	100	$0.11	5	62s
4	OpenAI / GPT 5.6 Lunacompare →	—	1	100	$0.04	5	38s
5	Anthropic / Sonnet 5compare →	—	1	0	$2.48	50	467s

Model comparison

each point is a model (cabin-pond) — hover for exact values

rated unrated

No rated models yet — cast decisive votes to plot the field.

Vote A/B Gallery Runsratings appear after first votes — BT fit pending

You are visitor number 004097 since ~~july 1999~~ july 2026

Best viewed with Netscape Navigator 4.07 at 800×600 · webmaster@blender-bench.org · sample data, ratings illustrative