- Peringkat
- #36
- Total token output
- 254,892
- Waktu respons (rata-rata)
- 11.67s
- Total Biaya
- $0.156
Muse Ember (high) vs Muse Wonder (medium)
Skor rata-rata hampir imbang di 9.2 vs 9.2. Muse Ember (high) memiliki biaya benchmark lebih rendah di $0.156 vs $3.397. Muse Ember (high) lebih cepat di 11.67s vs 21.58s, dengan tingkat keberhasilan 79.7% vs 76.8%.
Model yang Dibandingkan
- Peringkat
- #39
- Total token output
- 121,546
- Waktu respons (rata-rata)
- 21.58s
- Total Biaya
- $3.397
Model yang direkomendasikan
Muse Ember (high)
It has the best score here (9.2), while costing about 21.8x less than Muse Wonder (medium).
Perbandingan terperinci
| Metrik | Muse Ember Muse Ember high | Muse Wonder Muse Wonder medium |
|---|---|---|
| Skor | 9.2 | 9.2 |
| Peringkat | #36 | #39 |
| Keandalan | 10.0 | 9.1 |
| Konsistensi | 9.6 | 9.6 |
| Percobaan | 69/69 | 69/69 |
| Tes benar | ||
| Tingkat lulus per percobaan | 79.7% | 76.8% |
| Tes tidak stabil | 1 | 1 |
| Total Run | 69 | 69 |
| Biaya per hasil | 0.865 | 19.982 |
| Total Biaya | $0.156 | $3.397 |
| Harga input | $0.100 / 1M | $4.000 / 1M |
| Harga output | $0.500 / 1M | $20.000 / 1M |
| Harga baca cache | $0.004 / 1M | $0.200 / 1M |
| Harga tulis cache | $0.125 / 1M | T/A |
| Total token input | 282,192 | 241,489 |
| Token output | 14,054 | 11,575 |
| Token penalaran | 240,838 | 109,971 |
| Waktu respons (rata-rata) | 11.67s | 21.58s |
| Waktu respons (maks) | 53.01s | 162.73s |
| Waktu respons (total) | 268.45s | 496.40s |
| Parameter | ~150B total (~10B aktif) | ~1T total (~50B 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.
#36 Muse Ember
high- Biaya
- $0.005
- Waktu
- 18.3s
- Token
- 9,965 tok
#39 Muse Wonder
medium- Biaya
- $0.208
- Waktu
- 64.8s
- Token
- 10,444 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 |
|---|---|---|---|---|---|---|---|---|---|
| Muse Ember | 10.0 | 10.0 | 100.0% | 0 | 48.28s | 149,560 | 2,625 | 1,759 | |
| Muse Wonder | 10.0 | 10.0 | 100.0% | 0 | 46.05s | 148,242 | 2,406 | 2,850 |
| Trik anti-AI | Skor | Konsistensi | Tingkat lulus per percobaan | Tes tidak stabil | Tes benar | Waktu respons (rata-rata) | Token input | Token output | Token penalaran |
|---|---|---|---|---|---|---|---|---|---|
| Muse Ember | 10.0 | 10.0 | 100.0% | 0 | 2.29s | 798 | 275 | 5,440 | |
| Muse Wonder | 8.3 | 10.0 | 75.0% | 0 | 4.84s | 594 | 378 | 3,874 |
| Pemrograman | Skor | Konsistensi | Tingkat lulus per percobaan | Tes tidak stabil | Tes benar | Waktu respons (rata-rata) | Token input | Token output | Token penalaran |
|---|---|---|---|---|---|---|---|---|---|
| Muse Ember | 10.0 | 10.0 | 100.0% | 0 | 9.62s | 7,410 | 1,213 | 34,691 | |
| Muse Wonder | 10.0 | 10.0 | 100.0% | 0 | 26.87s | 7,401 | 1,181 | 26,424 |
| Gabungan | Skor | Konsistensi | Tingkat lulus per percobaan | Tes tidak stabil | Tes benar | Waktu respons (rata-rata) | Token input | Token output | Token penalaran |
|---|---|---|---|---|---|---|---|---|---|
| Muse Ember | 10.0 | 10.0 | 100.0% | 0 | 10.86s | 102,312 | 5,879 | 17,216 | |
| Muse Wonder | 10.0 | 10.0 | 100.0% | 0 | 28.23s | 66,033 | 4,934 | 17,166 |
| 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 |
|---|---|---|---|---|---|---|---|---|---|
| Muse Ember | 10.0 | 10.0 | 100.0% | 0 | 3.86s | 7,215 | 342 | 4,585 | |
| Muse Wonder | 10.0 | 10.0 | 100.0% | 0 | 6.12s | 7,053 | 433 | 3,670 |
| Spesifik domain | Skor | Konsistensi | Tingkat lulus per percobaan | Tes tidak stabil | Tes benar | Waktu respons (rata-rata) | Token input | Token output | Token penalaran |
|---|---|---|---|---|---|---|---|---|---|
| Muse Ember | 5.3 | 10.0 | 33.3% | 0 | 29.08s | 873 | 1,687 | 119,272 | |
| Muse Wonder | 5.3 | 10.0 | 33.3% | 0 | 70.72s | 555 | 960 | 42,167 |
| Kecerdasan umum | Skor | Konsistensi | Tingkat lulus per percobaan | Tes tidak stabil | Tes benar | Waktu respons (rata-rata) | Token input | Token output | Token penalaran |
|---|---|---|---|---|---|---|---|---|---|
| Muse Ember | 10.0 | 10.0 | 100.0% | 0 | 2.80s | 522 | 201 | 1,993 | |
| Muse Wonder | 10.0 | 10.0 | 100.0% | 0 | 7.32s | 519 | 147 | 1,123 |
| Kepatuhan instruksi | Skor | Konsistensi | Tingkat lulus per percobaan | Tes tidak stabil | Tes benar | Waktu respons (rata-rata) | Token input | Token output | Token penalaran |
|---|---|---|---|---|---|---|---|---|---|
| Muse Ember | 9.8 | 10.0 | 100.0% | 0 | 3.05s | 759 | 159 | 4,867 | |
| Muse Wonder | 6.6 | 10.0 | 50.0% | 0 | 3.88s | 523 | 124 | 1,783 |
| Pemecahan teka-teki | Skor | Konsistensi | Tingkat lulus per percobaan | Tes tidak stabil | Tes benar | Waktu respons (rata-rata) | Token input | Token output | Token penalaran |
|---|---|---|---|---|---|---|---|---|---|
| Muse Ember | 5.4 | 7.2 | 44.4% | 1 | 12.05s | 786 | 1,091 | 33,034 | |
| Muse Wonder | 7.7 | 10.0 | 66.7% | 0 | 9.63s | 564 | 481 | 4,834 |
| Pemanggilan alat | Skor | Konsistensi | Tingkat lulus per percobaan | Tes tidak stabil | Tes benar | Waktu respons (rata-rata) | Token input | Token output | Token penalaran |
|---|---|---|---|---|---|---|---|---|---|
| Muse Ember | 10.0 | 10.0 | 100.0% | 0 | 3.95s | 11,729 | 319 | 1,231 | |
| Muse Wonder | 10.0 | 10.0 | 100.0% | 0 | 9.26s | 9,780 | 369 | 2,803 |
| Pengetahuan umum | Skor | Konsistensi | Tingkat lulus per percobaan | Tes tidak stabil | Tes benar | Waktu respons (rata-rata) | Token input | Token output | Token penalaran |
|---|---|---|---|---|---|---|---|---|---|
| Muse Ember | 3.0 | 10.0 | 0.0% | 0 | 16.46s | 228 | 263 | 16,750 | |
| Muse Wonder | 4.7 | 1.6 | 66.7% | 1 | 16.33s | 225 | 162 | 3,277 |
Perbandingan Cepat
Ganti Pasangan Perbandingan
Seed 2.1 TurbolowvsMuse EmberhighClaude Sonnet 5.5maxvsMuse WondermediumDeepSeek V4.1 Flash 1017lowvsMuse WondermediumClaude Opus 5.5mediumvsMuse EmberhighMuse EmberhighvsGPT-5.5mediumSeed 2.1 TurbolowvsMuse WondermediumMuse WondermediumvsQwen3.8 Max (0902)highClaude Sonnet 5.5maxvsMuse EmberhighMuse EmberhighvsGPT-6.1 TerralowMuse EmberhighvsQwen3.8 MAXmediumClaude Haiku 5.5xhighvsMuse WondermediumDeepSeek V4.1 Flash 1017lowvsMuse Emberhigh