Text-to-Optimization on Text2Opt-Bench eval-subset 1.0 benchmark leaderboard · SOTA2 Research

Text-to-Optimization on Text2Opt-Bench eval-subset 1.0

89.9Resource Allocation ScoreOpus 3.5

Evaluation Results

Method	Links
Opus 3.52026.05		89.9	—	98	96	100	38	96	100	88	92	62	88	96	86.7
GPT-4o2026.05		87.9	—	100	86	90	70	88	100	98	80	70	70	96	86.2
Sonnet 3.52026.05		84.7	—	100	96	98	50	100	98	64	98	62	98	100	87.6
DS-R12026.05		80.6	—	100	78	96	34	34	94	64	70	60	76	96	72.9
o1-mini2026.05		80.2	—	100	94	96	34	82	98	70	76	66	68	100	80.4
DS-V32026.05		79	—	100	90	96	22	62	98	54	66	18	86	100	72
Llama3.3-70B2026.05		49.6	—	88	30	0	0	0	98	16	10	6	42	96	35.1
GPT-4o-mini2026.05		49.2	—	100	62	82	2	26	90	50	56	32	60	90	59.1
Qwen2.5-7B2026.05		13.3	—	38	0	0	0	0	6	0	0	0	4	2	4.5