Tool Use on Synthetic Data (test) benchmark leaderboard · SOTA2 Research

Tool Use on Synthetic Data (test)

92.29Task AccuracyRISE

Evaluation Results

Method	Links
RISE2026.01		92.29	97.84	58.24	69.13	86.83	81.63
Qwen2.5-7B-Instruct2026.01		92.28	97.7	55.98	67.9	85.73	80.16
RISE2026.01		91.98	98.81	62.19	50.15	85.44	82.66
RISE2026.01		90.9	95.11	60.62	44.6	84.26	84.23
RISE2026.01		89.93	81.24	50.86	60.79	79.36	75.81
deepseek-llm-7b-chat2026.01		89.81	81.28	48.43	60.34	77.39	74.66
Llama-3.1-8B-Instruct2026.01		88.58	98.33	60.02	47.22	83.2	79.94
Qwen3-8B2026.01		88.58	83.68	58.88	44.14	83.07	84.16
ToolLLaMA-2-7b-v22026.01		87.38	72.31	32.68	38.21	53.27	56.42
RISE2026.01		83.18	73.41	47.06	27.47	68.56	73.17
Mistral-7B2026.01		83.03	72.9	45.77	26.7	67.02	71.52
Mistral-7B2026.01		70.83	71.32	32.1	11.57	60.91	53.57
Mistral-7B2026.01		69.75	69.9	35.84	11.76	60.3	53.51
GPT-5-nano2026.01		68.99	92.94	57.13	29.61	59.39	79.4
Llama-3.1-8B-Instruct2026.01		68.83	63.08	38.64	8.33	73.22	47.99
Qwen3-8B2026.01		67.59	60.77	32.38	8.33	74.82	48.34
Qwen3-8B2026.01		65.9	63.34	47.06	24.38	56.8	58.78
Llama-3.1-8B-Instruct2026.01		65.6	62.09	32.38	8.33	74.82	48.34
Qwen2.5-7B-Instruct2026.01		60.03	62.73	48.33	41.67	64.1	65.73
Qwen2.5-7B-Instruct2026.01		58.49	61.92	49.68	39.66	62.73	63.94
deepseek-llm-7b-chat2026.01		50.93	21.44	20.85	7.56	13.12	15.1
deepseek-llm-7b-chat2026.01		47.84	24.65	17.45	5.86	18.08	21.81
ToolAlpaca-7B2026.01		41.27	63.94	9.63	17.18	51.3	46.3
Lynx-7b2026.01		6.35	94.26	1.75	3.51	22.36	26.9