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

Financial Backtesting Accuracy on BacktestBench synthetic (test)

67.41Overall Accuracy (OA)

Gemini 3 Pro

Updated 15d ago

Evaluation Results

Method	Links
Gemini 3 Pro 2026.05		67.41	51.67	90.37	82.46	89.07
Qwen3 Max 2026.05		61.84	44.14	85.56	82.28	85.27
Seed 1.8 2026.05		60.58	43.22	85.83	77.05	84.8
GLM 4.7 2026.05		56.83	39.13	80.33	77.24	80.76
Qwen3 235B 2026.05		55.01	34.71	82.26	75.93	85.04
Deepseek V3.2 2026.05		50.7	29.18	79.78	73.51	81.47
GPT OSS 120B 2026.05		49.56	30.86	74	70.34	76.48
Qwen3 32B 2026.05		47.81	27.93	75.93	70.34	72.21
Qwen3 30B 2026.05		47.03	28.67	71.25	65.86	75.06
MiniMax M2.1 2026.05		45.91	21.98	81.71	69.22	76.72
Kimi K2 Thinking 2026.05		44.97	30.02	66.57	58.02	67.46
Qwen3 Next 80B 2026.05		44.06	26.02	67.4	65.86	68.17
Qwen3 Coder Plus 2026.05		42.05	20.59	63.41	71.08	77.91
Mimo V2 Flash 2026.05		40.43	17.43	73.04	64.37	71.26
Qwen3 14B 2026.05		36.97	16.96	62.04	61.57	64.61
GPT OSS 20B 2026.05		33.13	17.33	51.17	49.44	62
Ministral 3 14B 2026.05		26.98	7.76	47.73	50.93	58.91
Qwen3 8B 2026.05		26.36	4.93	53.92	50.56	57.48
GLM 4.7 Flash 2026.05		20.26	4.41	47.59	37.31	32.3
Qwen3 4B 2026.05		19.63	1.77	39.06	42.54	48.22
Ministral 3 8B 2026.05		17.26	2.83	34.11	37.31	36.34
Seed OSS 36B 2026.05		11.34	4.6	20.08	13.06	28.5
Kimi Linear 48B 2026.05		4.87	0.19	10.18	8.58	14.96