- Peringkat
- #120
- Total token output
- 66,174
- Waktu respons (rata-rata)
- 7.71s
- Total Biaya
- $0.063
Claude Haiku 5.5 (medium) vs DeepSeek V4 Flash 0731 (medium)
DeepSeek V4 Flash 0731 (medium) unggul dalam skor rata-rata dengan 7.8 vs 7.7. Claude Haiku 5.5 (medium) memiliki biaya benchmark lebih rendah di $0.063 vs $0.460. Claude Haiku 5.5 (medium) lebih cepat di 7.71s vs 94.16s, dengan tingkat keberhasilan 71.0% vs 72.5%.
Model yang Dibandingkan
- Peringkat
- #117
- Total token output
- 358,225
- Waktu respons (rata-rata)
- 94.16s
- Total Biaya
- $0.460
Model yang direkomendasikan
Claude Haiku 5.5 (medium)
Its score stays close to the best score here (7.7 vs 7.8), while costing about 7.3x less than DeepSeek V4 Flash 0731 (medium).
Perbandingan terperinci
| Metrik | Claude Haiku 5.5 Claude Haiku 5.5 medium | DeepSeek V4 Flash 0731 DeepSeek V4 Flash 0731 medium |
|---|---|---|
| Skor | 7.7 | 7.8 |
| Peringkat | #120 | #117 |
| Keandalan | 10.0 | 10.0 |
| Konsistensi | 7.5 | 7.2 |
| Percobaan | 69/69 | 69/69 |
| Tes benar | ||
| Tingkat lulus per percobaan | 71.0% | 72.5% |
| Tes tidak stabil | 7 | 8 |
| Total Run | 69 | 69 |
| Biaya per hasil | 0.485 | 1.176 |
| Total Biaya | $0.063 | $0.460 |
| Harga input | $0.100 / 1M | $0.005 / 1M |
| Harga output | $0.500 / 1M | $1.280 / 1M |
| Harga baca cache | $0.010 / 1M | $0.005 / 1M |
| Harga tulis cache | $0.125 / 1M | T/A |
| Total token input | 298,334 | 274,189 |
| Token output | 10,328 | 66,689 |
| Token penalaran | 55,846 | 510,525 |
| Waktu respons (rata-rata) | 7.71s | 94.16s |
| Waktu respons (maks) | 47.70s | 303.26s |
| Waktu respons (total) | 177.29s | 2165.72s |
| Parameter | ~50B | 284B total (13B aktif) |
| Ketersediaan | Tertutup | Tertutup |
Harga cache berlaku untuk token input. Pembacaan memakai kembali prompt tersimpan; penulisan menyimpannya dan dapat menambah biaya. Token output menggunakan harga output.
Showcase generasi model
Hamster playing table tennis
Prompt: Create a detailed SVG illustration of a hamster playing table tennis.
#120 Claude Haiku 5.5
medium- Biaya
- $0.002
- Waktu
- 17.2s
- Token
- 3,524 tok
#117 DeepSeek V4 Flash 0731
medium- Biaya
- $0.004
- Waktu
- 93.3s
- Token
- 13,252 tok
Rincian Kategori
| Tugas agen | Skor | Konsistensi | Tingkat lulus per percobaan | Tes tidak stabil | Tes benar | Waktu respons (rata-rata) | Token input | Token output | Token penalaran |
|---|---|---|---|---|---|---|---|---|---|
| Claude Haiku 5.5 | 10.0 | 10.0 | 100.0% | 0 | 47.70s | 163,704 | 2,705 | 6,272 | |
| DeepSeek V4 Flash 0731 | 10.0 | 10.0 | 100.0% | 0 | 303.26s | 184,370 | 8,291 | 18,068 |
| Trik anti-AI | Skor | Konsistensi | Tingkat lulus per percobaan | Tes tidak stabil | Tes benar | Waktu respons (rata-rata) | Token input | Token output | Token penalaran |
|---|---|---|---|---|---|---|---|---|---|
| Claude Haiku 5.5 | 8.7 | 7.9 | 91.7% | 1 | 2.36s | 858 | 471 | 1,871 | |
| DeepSeek V4 Flash 0731 | 8.2 | 7.9 | 83.3% | 1 | 17.95s | 698 | 2,101 | 6,715 |
| Pemrograman | Skor | Konsistensi | Tingkat lulus per percobaan | Tes tidak stabil | Tes benar | Waktu respons (rata-rata) | Token input | Token output | Token penalaran |
|---|---|---|---|---|---|---|---|---|---|
| Claude Haiku 5.5 | 6.3 | 4.8 | 66.7% | 2 | 7.16s | 10,608 | 518 | 11,415 | |
| DeepSeek V4 Flash 0731 | 6.4 | 4.4 | 77.8% | 2 | 198.78s | 6,032 | 4,109 | 160,154 |
| Gabungan | Skor | Konsistensi | Tingkat lulus per percobaan | Tes tidak stabil | Tes benar | Waktu respons (rata-rata) | Token input | Token output | Token penalaran |
|---|---|---|---|---|---|---|---|---|---|
| Claude Haiku 5.5 | 7.3 | 5.8 | 83.3% | 1 | 20.29s | 97,091 | 4,812 | 17,244 | |
| DeepSeek V4 Flash 0731 | 7.3 | 5.8 | 83.3% | 1 | 150.58s | 65,118 | 40,110 | 97,296 |
| Parsing dan ekstraksi data | Skor | Konsistensi | Tingkat lulus per percobaan | Tes tidak stabil | Tes benar | Waktu respons (rata-rata) | Token input | Token output | Token penalaran |
|---|---|---|---|---|---|---|---|---|---|
| Claude Haiku 5.5 | 10.0 | 10.0 | 100.0% | 0 | 3.43s | 10,515 | 312 | 1,666 | |
| DeepSeek V4 Flash 0731 | 10.0 | 10.0 | 100.0% | 0 | 5.21s | 7,290 | 287 | 1,519 |
| Spesifik domain | Skor | Konsistensi | Tingkat lulus per percobaan | Tes tidak stabil | Tes benar | Waktu respons (rata-rata) | Token input | Token output | Token penalaran |
|---|---|---|---|---|---|---|---|---|---|
| Claude Haiku 5.5 | 2.9 | 4.4 | 22.2% | 2 | 7.96s | 990 | 71 | 10,376 | |
| DeepSeek V4 Flash 0731 | 2.9 | 7.2 | 11.1% | 1 | 203.11s | 592 | 9,333 | 154,069 |
| Kecerdasan umum | Skor | Konsistensi | Tingkat lulus per percobaan | Tes tidak stabil | Tes benar | Waktu respons (rata-rata) | Token input | Token output | Token penalaran |
|---|---|---|---|---|---|---|---|---|---|
| Claude Haiku 5.5 | 3.8 | 2.5 | 33.3% | 1 | 5.17s | 714 | 192 | 1,933 | |
| DeepSeek V4 Flash 0731 | 6.1 | 3.1 | 66.7% | 1 | 12.16s | 471 | 66 | 1,589 |
| Kepatuhan instruksi | Skor | Konsistensi | Tingkat lulus per percobaan | Tes tidak stabil | Tes benar | Waktu respons (rata-rata) | Token input | Token output | Token penalaran |
|---|---|---|---|---|---|---|---|---|---|
| Claude Haiku 5.5 | 10.0 | 10.0 | 100.0% | 0 | 3.01s | 921 | 100 | 1,278 | |
| DeepSeek V4 Flash 0731 | 8.5 | 6.8 | 83.3% | 1 | 2.42s | 706 | 109 | 580 |
| Pemecahan teka-teki | Skor | Konsistensi | Tingkat lulus per percobaan | Tes tidak stabil | Tes benar | Waktu respons (rata-rata) | Token input | Token output | Token penalaran |
|---|---|---|---|---|---|---|---|---|---|
| Claude Haiku 5.5 | 7.7 | 10.0 | 66.7% | 0 | 2.67s | 912 | 568 | 2,119 | |
| DeepSeek V4 Flash 0731 | 8.2 | 7.3 | 88.9% | 1 | 65.12s | 594 | 1,530 | 66,970 |
| Pemanggilan alat | Skor | Konsistensi | Tingkat lulus per percobaan | Tes tidak stabil | Tes benar | Waktu respons (rata-rata) | Token input | Token output | Token penalaran |
|---|---|---|---|---|---|---|---|---|---|
| Claude Haiku 5.5 | 10.0 | 10.0 | 100.0% | 0 | 5.02s | 11,757 | 354 | 609 | |
| DeepSeek V4 Flash 0731 | 10.0 | 10.0 | 100.0% | 0 | 4.41s | 8,178 | 374 | 152 |
| Pengetahuan umum | Skor | Konsistensi | Tingkat lulus per percobaan | Tes tidak stabil | Tes benar | Waktu respons (rata-rata) | Token input | Token output | Token penalaran |
|---|---|---|---|---|---|---|---|---|---|
| Claude Haiku 5.5 | 3.0 | 10.0 | 0.0% | 0 | 3.12s | 264 | 225 | 1,063 | |
| DeepSeek V4 Flash 0731 | 3.0 | 10.0 | 0.0% | 0 | 56.64s | 140 | 379 | 3,413 |
Perbandingan Cepat
Ganti Pasangan Perbandingan
DeepSeek V4 Flash 0731mediumvsGrok 4.5lowClaude Haiku 5.5mediumvsMuse Glimmer 30BhighClaude Haiku 5.5mediumvsQwen3.7 FlashlowDeepSeek V4 Flash 0731mediumvsQwen3.7 PlusnoneClaude Haiku 5.5mediumvsGPT-6 LunahighDeepSeek V4 Flash 0731mediumvsGLM 5.3highDeepSeek V4 Flash 0731mediumvsGPT-6 LunahighDeepSeek V4 Flash 0731mediumvsMuse Glimmer 30BhighClaude Haiku 5.5mediumvsGrok 4.5lowDeepSeek V4 Flash 0731mediumvsLing 3.1 FlashhighDeepSeek V4 Flash 0731mediumvsQwen3.7 FlashlowClaude Haiku 5.5mediumvsQwen3.7 Plusnone