Pest Classification on AgriInsect (test) benchmark leaderboard · SOTA2 Research

Pest Classification on AgriInsect (test)

91.24Beetle AccuracyGemini-2.5-Flash

Evaluation Results

Method	Links
Gemini-2.5-Flash2026.05		91.24	77.08	84.94	93.02	70.91	50	87.14	61.11	60	82.05
GPT-52026.05		87.42	84	83.59	100	81.03	64.71	85.62	44.44	54.55	83.24
Pest-Thinker-7B2026.05		84.77	66	86.33	95.74	81.9	70.59	87.58	88.89	54.55	82.77
GPT-5-mini2026.05		82.12	72	69.92	85.11	68.97	47.06	78.43	50	54.55	73.34
Pest-Thinker-SFT-7B2026.05		76.75	60	70.31	76.77	79.31	69.31	74.12	78.89	45.45	73
Pest-Thinker-3B2026.05		69.54	62	77.73	85.11	77.59	58.82	79.74	55.56	61.11	73.92
Qwen3VL-30B-Thinking2026.05		67.67	53.49	61.44	74.42	51.38	23.08	70.83	57.14	85.71	61.97
Qwen2.5VL-72B-Instruct2026.05		64.67	44.68	56.47	82.98	50.86	41.18	64.43	55.56	54.55	59.06
Qwen2.5VL-32B-Instruct2026.05		60.93	54	46.88	72.34	38.79	41.18	66.01	50	27.27	52.62
Pest-Thinker-SFT-3B2026.05		60.26	62	66.8	77.23	61.21	41.18	67.97	55.56	18.18	64.14
Qwen3VL-8B-Thinking2026.05		59.59	47.62	46.09	60.87	43.64	35.29	55.32	47.06	55.56	50.51
VisionThink-Efficient-7B2026.05		58.94	44	50.78	68.09	49.14	17.65	64.05	55.56	27.27	53.78
VisionThink-General-7B2026.05		58.28	40	50.78	74.47	52.59	35.29	64.71	44.44	36.36	54.6
Vision-SR1-7B2026.05		56.29	46	48.05	68.09	48.28	41.18	59.48	55.56	36.36	51.92
Qwen3VL-30B-Instruct2026.05		56.08	68	53.54	85.11	49.14	35.29	64.24	55.56	45.45	56.81
Vision-G1-7B2026.05		54.97	38	50.78	68.09	53.45	29.41	60.13	44.44	36.36	52.74
Qwen2.5VL-7B-Instruct2026.05		51.66	34	51.95	57.45	43.97	17.65	55.56	33.33	54.55	49.48
Qwen3VL-8B-Instruct2026.05		51.35	42.55	40.71	77.78	39.13	17.65	44.22	50	33.33	44.68
Qwen2.5-VL-3B-Instruct2026.05		45.7	32	39.84	63.83	40.52	11.76	43.79	38.89	18.18	42.14
InternVL3-8B2026.05		41.06	48	39.84	51.06	50	11.76	46.41	22.22	27.27	42.03
InternVL3.5-8B2026.05		37.75	30	37.5	42.55	32.76	35.29	35.95	11.11	18.18	35.51