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

Time Series Reasoning on TRQA

83.34Accuracy

ARTIST

Updated 1mo ago

Evaluation Results

Method	Links
ARTIST 2026.02		83.34	-
ARTIST 2026.02		83.06	78.02
ARTIST 2026.02		82.26	62.32
ARTIST 2026.02		82.26	-
T2SP 2026.06		81.5	89.8
ITFormer-4B 2026.02		80.12	74.22
OpenTSLM-4B 2026.02		76.25	69.36
T2SP 2026.06		76	86.4
T2SP 2026.06		76	86.4
Qwen-3.5-9B 2026.06		75.5	86
Claude-haiku-4.5 2026.06		75	85.7
GPT-5.4 2026.06		74	85.1
Qwen-3.5-9B 2026.06		73.5	84.7
Gemini-3.1-flash-lite 2026.06		73	84.4
T2SP 2026.06		72.7	84.2
Gemini-3.1-flash-lite 2026.06		72.5	71.9
TimeMaster 2026.02		72.08	-
GPT-5.4 2026.06		71	83
Claude-haiku-4.5 2026.06		70	82.4
ChatTS-14B 2026.02		69	55.57
TimeMaster 2026.02		68.45	-
Base Model 2026.02		66.09	-
VL-Time 2026.02		58.5	-
Llama-3 8B 2026.02		57.69	53.88
ChatTS-14B 2026.02		57.5	42.12
Llama-3 8B 2026.02		55.5	53.02
VL-Time 2026.02		53.5	-
Random Guess 2026.02		37.13	37.13
Random Guess 2026.06		36.9	32.2
Qwen3-14B 2026.02		33	29.54
ChatTime-7B 2026.06		31.4	24.5
Qwen3-14B 2026.02		29	30.32
GPT-5 2026.02		25	28.7
GPT-5 2026.02		21.5	21.67
Qwen3-8B 2026.02		15.25	14.02
Qwen3-8B 2026.02		14.63	15.2