Share your thoughts, 1 month free Claude Pro on usSee more

Meme Generation on GOAT-G (test)

60.1BERTScore F1

Gemini-3-Flash

Updated 1mo ago

Evaluation Results

Method	Links
Gemini-3-Flash 2026.06		60.1	43.7
Claude-Sonnet-4.6 2026.06		58	41.2
Intent Projection 2026.06		53.1	33.8
GPT-5.4-mini 2026.06		52.4	33.8
Intent Projection 2026.06		52	32.3
InternVL3-8B 2026.06		51.2	31.9
Qwen3-VL-8B 2026.06		50.2	30.4
InternVL3-8B 2026.06		49.6	30.3
Qwen3-VL-8B 2026.06		48.6	28.9
InternVL3-8B 2026.06		48.3	29.2
Intent Projection 2026.06		47.4	28.3
Qwen3-VL-8B 2026.06		47.4	27.8
LLaVA-OneVision-7B 2026.06		45.6	26.6
InternVL3-8B 2026.06		45.5	26
Qwen3-VL-8B 2026.06		44	24.6
LLaVA-OneVision-7B 2026.06		43.9	25.1
InternVL3-8B 2026.06		43.9	24.1
LLaVA-OneVision-7B 2026.06		42.6	24
Qwen3-VL-8B 2026.06		42.1	22.5
LLaVA-OneVision-7B 2026.06		39.4	20.8
LLaVA-OneVision-7B 2026.06		37.9	19.1