#78 DeepSeek V4 Pro
none
Reached the allocated time limit (300 seconds) without receiving showcase output.
- Gharama
- $0.000
- Muda
- 300.0s
- Tokeni
- 0 tok
DeepSeek
Toleo: 2026-04-24
Imepimwa tarehe: 2026-04-29 14:46
deepseek/deepseek-v4-pro::none
1.6T jumla (49B vinavyotumika)MoEChanzo huria
(high)
(none)
Muhtasari
DeepSeek V4 Pro hupata alama 6.2 kwenye AI BENCHY na iko nafasi ya #78. Ina reliability 7.9, pass rate 48.2%, gharama jumla $0.043, na wastani wa response time 14.01s.
Kinachofanya DeepSeek V4 Pro iwe ya kipekee: Gharama yake ya benchmark ni ya chini isivyo kawaida kwa kiwango hiki cha score.
Taarifa za modeli
Ilifanyiwa utafiti tarehe 2026-08-12
- Vigezo
- 1.6T jumla (49B vinavyotumika)
- Muundo
- MoE
- Upatikanaji
- Chanzo huria
- Leseni
- MIT
6.2
Uthabiti
8.7
7.9
Jumla ya tokeni za matokeo
3,903
Jumla ya tokeni za ingizo
0
Bei ya ingizo
$0.435 / 1M
Bei ya toleo
$0.870 / 1M
Majaribio yasiyo thabiti
3
Majaribio yasiyo thabiti yalikuwa na matokeo mchanganyiko kati ya run (angalau kupita moja na kufeli moja).
Hamster playing table tennis
Prompt: Create a detailed SVG illustration of a hamster playing table tennis.
Historia ya uendeshaji
| Imepimwa tarehe | Alama | Uaminifu | Majaribio sahihi | Jumla ya gharama | Linganisha |
|---|---|---|---|---|---|
| 2026-08-14 03:00 Jaribio tena | 6.8 | 10.0 | $0.256 ↑ | Linganisha | |
| 2026-07-16 23:18 Jaribio jipya limeongezwa | 6.9 | 10.0 | $0.096 | Linganisha | |
| 2026-06-16 15:17 Jaribio tena | 7.2 | 9.9 | $0.034 | Linganisha | |
| 2026-06-16 14:39 Suite imebadilika | 7.2 | 9.9 | $0.030 | Linganisha | |
| 2026-06-04 14:24 Jaribio jipya limeongezwa | 5.7 | 8.5 | $0.025 ↓ | Linganisha | |
| 2026-05-22 00:38 Suite imebadilika | 6.0 | 8.1 | $0.046 | Linganisha | |
| 2026-04-29 14:46 Jaribio tena | 6.2 | 7.9 | $0.043 | Uendeshaji wa sasa | |
| 2026-04-24 09:19 Uendeshaji wa kwanza | 3.1 | Haipo | $0.009 | Linganisha |
Ulinganisho wa uendeshaji
| Uendeshaji | Ufunikaji wa benchmark | Alama | Uthabiti | Uaminifu | Majaribio sahihi | Majaribio yasiyo thabiti | Jumla ya tokeni za matokeo | Jumla ya tokeni za ingizo | Jumla ya gharama | Muda wa majibu (wastani) |
|---|---|---|---|---|---|---|---|---|---|---|
| 2026-04-29 14:46 · Jaribio tena | marudio 54/54 | 6.2 | 8.7 | 7.9 | 7/18 | 3 | 3,903 | 0 | $0.043 | 14.01s |
| 2026-06-04 14:24 · Jaribio jipya limeongezwa | marudio 63/63 | 5.7 | 8.5 | 8.5 | 7/21 | 4 | 5,349 | 44,845 | $0.025 | 12.38s |
| Tofauti | — | +0.5 | +0.2 | -0.6 | 0 | -1 | -1446 | -44845 | +$0.018 | +1623ms |
Ufunikaji wa benchmark unatofautiana: marudio 54/54 (Lengo: marudio 3 kwa kila jaribio) dhidi ya marudio 63/63 (Lengo: marudio 3 kwa kila jaribio). Jumla na vipimo vinavyoathiriwa na marudio haviwezi kulinganishwa moja kwa moja.
Uendeshaji huu miwili ulitumia suite tofauti za benchmark, hivyo tofauti zinaonyesha mabadiliko ya modeli na ya suite pia.
Chati
Chagua modeli ya kwanza, kisha bofya modeli ya pili kufungua ukurasa wa kulinganisha bega kwa bega.
Modeli bora kwa alama
Alama vs Jumla ya gharama
Muda wa majibu (wastani)
Alama vs Muda wa majibu (wastani)
Jumla ya tokeni za matokeo
Alama vs Jumla ya tokeni za matokeo
Ulinganisho wa haraka
DeepSeek V4 PrononevsSolar Pro 4mediumDeepSeek V4 PrononevsGLM 5.3lowDeepSeek V4 PrononevsSolar Pro 4lowDeepSeek V4 PrononevsMiMo-V2.5-PromediumDeepSeek V4 PrononevsSeed-2.0-CodemediumDeepSeek V4 PrononevsGemini 3 Flash PreviewnoneDeepSeek V4 PrononevsStep 3.7 FlashhighDeepSeek V4 PrononevsGemma 4 26B A4BmediumInapatikana bureDeepSeek V4 PrononevsQwen3.6 FlashmediumDeepSeek V4 PrononevsGLM 5.2none
Mgawanyo wa kategoria
| Kategoria | Alama | Uthabiti | Majaribio sahihi |
|---|---|---|---|
| Mbinu za kupinga AI | 3.5 | 8.0 | |
| Uandishi wa msimbo | 7.1 | 3.7 | |
| Mchanganyiko | 9.5 | 10.0 | |
| Uchanganuzi na uchimbaji wa data | 10.0 | 10.0 | |
| Mahususi kwa domeni | 5.3 | 10.0 | |
| Akili ya jumla | 4.3 | 9.9 | |
| Ufuataji wa maagizo | 6.3 | 10.0 | |
| Utatuzi wa mafumbo | 6.0 | 7.1 | |
| Mwito wa zana | 10.0 | 10.0 |