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

Tool-use task completion on SkillCraft Overall

0.26Avg Tokens (M)

GPT-5.2

Updated 1mo ago

Evaluation Results

Method	Links
GPT-5.2 2026.02		0.26	84	3.8	0.43	9.9	8.9	114
Kimi-K2-Thinking 2026.02		0.3	70	3.4	0.13	8.3	11.9	56
Gemini 3 Pro 2026.02		0.3	93	3.9	0.3	11.9	11.9	116
Minimax-M2.1 2026.02		0.38	100	3.2	0.04	5.2	15.4	119
Claude 4.5 Sonnet 2026.02		0.4	81	3.4	0.28	10.2	9.2	121
DeepSeek-R1 2026.02		0.41	62	3.4	0.18	9.9	11.7	101
Minimax-M2.1 2026.02		0.42	-	-	0.04	5.5	16.6	117
GLM-4.7 2026.02		0.48	91	3.7	0.12	13	13.3	108
Kimi-K2-Thinking 2026.02		0.51	-	-	0.21	6.7	16.8	55
DeepSeek-V3.2-EXP 2026.02		0.53	71	4.8	0.1	15.4	14.9	87
DeepSeek-R1 2026.02		0.58	-	-	0.24	9	13.4	89
Gemini 3 Pro 2026.02		0.66	-	-	0.59	10.5	20.2	108
GLM-4.7 2026.02		0.78	-	-	0.2	13.5	16.9	91
DeepSeek-V3.2-EXP 2026.02		1.04	-	-	0.21	18.8	19.2	76
GPT-5.2 2026.02		1.23	-	-	1.77	20.6	19.4	109
Claude 4.5 Sonnet 2026.02		1.36	-	-	1.08	15.3	14.3	119