AI BENCHY
Advertise here

Kategoria ya AI BENCHY

Orodha ya Mwito wa zana

Ona ni modeli gani za AI zinafanya vizuri zaidi katika Mwito wa zana, zipi zinabaki thabiti, na pengo kubwa liko wapi. Panga kwa: Kipimo ↑.

Modeli zilizoonyeshwa

15

Wastani wa Alama ya Mwito wa zana

8.7

Modeli bora

Grok 4.1 Fast 2.8

Sababu za kushindwa

Kwa sababu ya kushindwa Hitilafu ya API15 Kwa sababu ya kushindwa Mwito wa zana si sahihi7 Kwa sababu ya kushindwa Hakufuata maelekezo6 Kwa sababu ya kushindwa Hakuna jibu2 Kwa sababu ya kushindwa Jibu lisilo sahihi2
Nafasi Modeli Kampuni Alama ya Mwito wa zana Alama Majaribio sahihi Muda wa majibu (wastani)
#12 Gemini 3.1 Flash Lite Preview high Google 10.0 8.6 1/1 7.73s
#14 Qwen3.6 Max Preview medium Qwen 10.0 8.5 1/1 18.3s
#15 GPT-5.3-Codex medium OpenAI 10.0 8.4 1/1 6.37s
#16 Gemini 3 Flash Preview low Google 10.0 8.4 1/1 4.99s
#17 GLM 5 medium Z.ai 10.0 8.3 1/1 15.9s
#18 Qwen3.7 Plus medium Qwen 10.0 8.2 1/1 15.0s
#19 Seed-2.0-Lite medium Bytedance Seed 10.0 8.2 1/1 12.4s
#21 GPT-5.4 medium OpenAI 10.0 8.0 1/1 13.3s
#22 Step 3.7 Flash medium Stepfun 10.0 8.0 1/1 4.16s
#23 GLM 5 Turbo medium Z.ai 10.0 8.0 1/1 9.84s
#24 GPT-5.2 Chat none OpenAI 10.0 7.9 1/1 4.68s
#25 Qwen3.5 Plus 2026-02-15 medium Qwen 10.0 7.9 1/1 7.54s
#26 Qwen3.6 Plus medium Qwen 10.0 7.9 1/1 5.87s
#28 Gemini 2.5 Flash medium Google 10.0 7.8 1/1 6.20s
#29 Qwen3.5-122B-A10B medium Qwen 10.0 7.8 1/1 4.60s

Modeli bora kwa Alama ya Mwito wa zana

Alama ya Mwito wa zana dhidi ya jumla ya gharama

Modeli bora kwa Muda wa majibu (wastani)