#7
GPT-5.4
OpenAI · Toleo: 2026-03-05 · openai/gpt-5.4::medium
Wastani wa alama
8.2
Gharama kwa matokeo
6.533
Uthabiti
8.9
Jumla ya gharama
$0.784
Majaribio sahihi
12
Jaribio huhesabiwa kuwa limepita kikamilifu tu ikiwa run zake zote zimepita.
Majaribio yenye makosa
3
Kiwango cha kupita kwa kila jaribio: 86.7%
Majaribio yasiyo thabiti
2
Majaribio yasiyo thabiti yalikuwa na matokeo mchanganyiko kati ya run (angalau kupita moja na kufeli moja).
Muda wa majibu (wastani)
21.06s
Muda wa majibu (upeo): 100.41s
Muda wa majibu (jumla): 315.95s
Jibu lisilo sahihi: 2 Hakufuata maelekezo: 1
Modeli bora kwa alama
Chagua modeli ya kwanza, kisha bofya modeli ya pili kufungua ukurasa wa kulinganisha bega kwa bega.
Ulinganisho wa haraka
GPT-5.4mediumvsQwen3.5-122B-A10BmediumGPT-5.4mediumvsGemini 3 Flash PreviewlowGPT-5.4mediumvsQwen3.5-27BmediumGPT-5.4mediumvsGemini 3 Pro PreviewmediumGPT-5.4mediumvsGPT-5.3-CodexmediumGPT-5.4mediumvsGemini 3.1 Flash Lite PreviewhighGPT-5.4mediumvsGemini 3 Flash PreviewmediumGPT-5.4mediumvsGemini 3.1 Pro PreviewmediumGPT-5.4mediumvsStep 3.5 FlashmediumInapatikana bure
Mgawanyo wa kategoria
| Kategoria | Wastani wa alama | Uthabiti | Majaribio sahihi |
|---|---|---|---|
| Anti-AI Tricks | 10.0 | 10.0 | 3/3 |
| Combined | 10.0 | 10.0 | 1/1 |
| Data parsing and extraction | 9.9 | 10.0 | 2/2 |
| Domain specific | 4.0 | 7.2 | 1/3 |
| Instructions following | 10.0 | 10.0 | 2/2 |
| Puzzle Solving | 7.0 | 7.2 | 2/3 |
| Tool Calling | 10.0 | 10.0 | 1/1 |