Previous photoPickle SandwichNext photoPanini
Human 92.8% yes7.2% no Model average 99.6% yes0.4% no Most aligned model meta-llama/llama-3.2-11b-vision-instruct 
meta-llama/llama-3.2-11b-vision-instructLeast aligned models 29-way tie 
openai/o3-pro
moonshotai/kimi-k2.5
x-ai/grok-4-fast+26 moreHuman distribution 92.8% yes, 7.2% no over 656 explicit votes. Model average distribution 99.6% yes, 0.4% no across the current model set. Closest current model 91.4% yes. Least aligned models 7.2 point gap. Legacy GPT-4o baseline 100.0% yes with a 7.2 point gap against humans. Biggest model gap 7.2 percentage points on this image. Current classification People mostly said yes meta-llama/llama-3.2-11b-vision-instruct 
qwen/qwen2.5-vl-32b-instruct
amazon/nova-lite-v1amazon/nova-pro-v1 
anthropic/claude-opus-4.6anthropic/claude-sonnet-4.6 
baidu/ernie-4.5-vl-28b-a3b
google/gemini-2.5-progoogle/gemini-3-flash-preview google/gemini-3.1-pro-preview google/gemma-3-12b-it google/gemma-3-27b-it meta-llama/llama-4-maverick meta-llama/llama-4-scout 
minimax/minimax-01
mistralai/pixtral-large-2411moonshotai/kimi-k2.5 openai/gpt-4.1 openai/gpt-4.1-mini openai/gpt-4o openai/gpt-4o-2024-11-20 openai/gpt-4o-mini openai/gpt-5.4 openai/gpt-5.4-pro openai/o3 openai/o3-pro qwen/qwen-2-vl-72b-instruct qwen/qwen2.5-vl-72b-instruct qwen/qwen3.5-397b-a17b x-ai/grok-4-fast 
z-ai/glm-4.6v
TEAPeople mostly said yes
Benchmark image 12
Avocado Tea
Spicy Avocado Egg Salad "Sandwich"
A neat avocado-and-egg-salad tea sandwich looks like it was served beside very expensive gossip. It is obviously a sandwich, just one that speaks in a quieter accent than the rest of the dataset.
Under development: this benchmark and its published results are provisional, not final.
meta-llama/llama-3.2-11b-vision-instructopenai/o3-promoonshotai/kimi-k2.5x-ai/grok-4-fast+26 moreAt a glance
How this photo split the room
meta-llama/llama-3.2-11b-vision-instruct
29-way tie
Why this page matters
This is the compact read on where humans, models, and comments start disagreeing about the same image.
Model spread
How models line up against the crowd
Bars run from more no on the left to more yes on the right. The marker shows the human yes rate.
meta-llama/llama-3.2-11b-vision-instructqwen/qwen2.5-vl-32b-instructamazon/nova-lite-v1amazon/nova-pro-v1anthropic/claude-opus-4.6anthropic/claude-sonnet-4.6baidu/ernie-4.5-vl-28b-a3bgoogle/gemini-2.5-progoogle/gemini-3-flash-previewgoogle/gemini-3.1-pro-previewgoogle/gemma-3-12b-itgoogle/gemma-3-27b-itmeta-llama/llama-4-maverickmeta-llama/llama-4-scoutminimax/minimax-01mistralai/pixtral-large-2411moonshotai/kimi-k2.5openai/gpt-4.1openai/gpt-4.1-miniopenai/gpt-4oopenai/gpt-4o-2024-11-20openai/gpt-4o-miniopenai/gpt-5.4openai/gpt-5.4-proopenai/o3openai/o3-proqwen/qwen-2-vl-72b-instructqwen/qwen2.5-vl-72b-instructqwen/qwen3.5-397b-a17bx-ai/grok-4-fastz-ai/glm-4.6vVote card
Generated summary for this photo
Selected human comments
meta-llama/llama-3.2-11b-vision-instruct comments
google/gemini-3-flash-preview comments