#142 DeepSeek V4 Flash
none- Biaya
- $0.004
- Waktu
- 157.6s
- Token
- 11,297 tok
AI BENCHY Compare
Ringkasan
DeepSeek V4 Flash unggul dalam skor rata-rata dengan 5.5 vs 4.9. DeepSeek V4 Flash memiliki biaya benchmark lebih rendah di $0.007 vs $0.037. KAT-Coder-Air V2.5 lebih cepat di 6.99s vs 26.75s, dengan tingkat keberhasilan 30.2% vs 38.1%.
Benchmark dihasilkan dari suite pengujian AI BENCHY pada: 2026-07-14
| Metrik | DeepSeek V4 Flash DeepSeek V4 Flash none | KAT-Coder-Air V2.5 KAT-Coder-Air V2.5 none |
|---|---|---|
| Skor | 5.5 | 4.9 |
| Peringkat | #142 | #168 |
| Keandalan | 10.0 | 10.0 |
| Konsistensi | 8.9 | 7.1 |
| Tes benar | ||
| Tingkat lulus per percobaan | 30.2% | 38.1% |
| Tes tidak stabil | 3 | 7 |
| Total Run | 63 | 63 |
| Biaya per hasil | 0.203 | 0.724 |
| Total Biaya | $0.007 | $0.037 |
| Harga input | $0.090 / 1M | $0.150 / 1M |
| Harga output | $0.180 / 1M | $0.600 / 1M |
| Total token input | 50,127 | 41,328 |
| Token output | 13,710 | 49,933 |
| Token penalaran | 0 | 0 |
| Waktu respons (rata-rata) | 26.75s | 6.99s |
| Waktu respons (maks) | 111.96s | 24.94s |
| Waktu respons (total) | 561.82s | 146.78s |
Showcase generasi
Prompt: Create a detailed SVG illustration of a hamster playing table tennis.
| Pemrograman | Skor | Konsistensi | Tingkat lulus per percobaan | Tes tidak stabil | Tes benar | Waktu respons (rata-rata) | Token input | Token output | Token penalaran |
|---|---|---|---|---|---|---|---|---|---|
| DeepSeek V4 Flash | 4.2 | 7.4 | 11.1% | 1 | 17.13s | 7,279 | 9,717 | 0 | |
| KAT-Coder-Air V2.5 | 3.3 | 9.6 | 0.0% | 0 | 14.31s | 6,472 | 16,918 | 0 |
Ganti Pasangan Perbandingan