Memory-Augmented Dialogue on PersonaMem v1.0 (test) benchmark leaderboard · SOTA2 Research

Memory-Augmented Dialogue on PersonaMem v1.0 (test)

74.53Overall ScorePersonaTree-ALL

Evaluation Results

Method	Links
PersonaTree-ALL2026.01		74.53	81.4	70.91	26.88	92.93	82.01	87.72	82.35	—
PersonaTree-ALL2026.01		73.34	88.37	69.09	31.18	91.92	69.78	87.72	76.47	—
FluxMem2026.02		72.43	85.27	70.91	36.56	91.92	65.47	—	—	82.46
PersonaTree-ALL2026.01		70.8	88.37	63.64	24.73	86.87	68.35	91.23	70.59	—
ALL Dialogue2026.01		64.86	69.77	63.64	11.83	84.85	73.38	84.21	70.59	—
MemoryOS2026.01		64.52	79.84	65.45	16.13	87.88	64.75	66.67	64.71	—
ALL Dialogue2026.01		63.84	78.29	61.82	12.9	81.82	65.47	77.19	76.47	—
O-Mem2026.02		63.25	67.81	65.45	21.51	89.9	61.15	—	—	73.68
MemoryOS2026.01		61.97	75.19	67.27	9.68	80.81	63.31	73.68	70.59	—
MemoryOS2026.01		61.97	64.34	76.36	11.83	82.83	66.19	73.68	76.47	—
Mem02026.01		61.8	82.95	47.27	7.53	83.84	63.31	71.93	70.59	—
ALL Dialogue2026.01		61.8	79.07	63.64	15.05	87.88	53.96	68.42	70.59	—
A-Mem2026.01		60.61	76.74	43.64	6.45	85.86	64.03	71.93	76.47	—
A-Mem2026.01		60.44	80.62	50.91	8.6	85.86	66.19	47.37	70.59	—
Mem02026.01		60.27	77.52	54.55	9.68	85.86	64.03	50.88	76.47	—
Memory OS2026.02		59.97	72.72	72.72	17.2	78.79	58.27	—	—	56.14
A-Mem2026.02		59.75	63.01	69.09	27.96	85.86	54.68	—	—	57.89
LangMem2026.01		57.39	75.97	50.91	10.75	79.8	63.31	45.61	52.94	—
LangMem2026.01		56.54	62.79	54.55	18.28	78.79	66.19	49.12	41.18	—
LangMem2026.01		56.37	72.87	52.73	9.68	78.79	61.15	50.88	47.06	—
Only LLM2026.01		54.33	62.02	43.64	8.6	84.85	71.22	28.07	52.94	—
Only LLM2026.01		52.63	59.69	43.64	5.38	80.81	65.47	42.11	52.94	—
HippoRAG 22026.02		51.87	76.74	23.64	33.33	78.79	56.12	—	—	38.6
Mem02026.02		48.55	32.13	52.73	15.05	80.81	54.68	—	—	57.89
Mem02026.01		47.71	32.56	52.73	21.51	73.74	57.55	47.37	58.82	—
A-Mem2026.01		47.03	27.13	45.45	27.96	76.77	55.4	43.86	76.47	—
Only LLM2026.01		44.14	30.23	38.18	19.35	76.77	55.4	38.6	41.18	—
LangMem2026.02		40.64	31.29	40	24.73	81.82	53.24	—	—	8.77