- Alama
- 8.6 (alama +0.41)
- Jumla ya gharama (bei ya sasa)
- $0.061 · nafuu kwa 96.3%
- Muda wa majibu (wastani)
- 7.51s · haraka kwa 86.2%
Moonshot AI
Toleo: 2026-10-24
Imepimwa tarehe: 2026-10-24 20:16
moonshotai/kimi-k3.1-pro::high
~2.8T jumla (~104B vinavyotumika)MoEImefungwaMakadirio
Muhtasari
Kimi K3.1 Pro hupata alama 8.2 kwenye AI BENCHY na iko nafasi ya #107. Ina reliability 8.5, pass rate 79.7%, gharama jumla $1.646, na wastani wa response time 54.31s.
Modeli mbadala zinazopendekezwa
Tunapendekeza uchague mojawapo ya modeli hizi badala yake:
- Alama
- 9.4 (alama +1.20)
- Jumla ya gharama (bei ya sasa)
- $0.183 · nafuu kwa 88.9%
- Muda wa majibu (wastani)
- 6.24s · haraka kwa 88.5%
- Alama
- 8.3 (alama +0.10)
- Jumla ya gharama (bei ya sasa)
- $0.047 · nafuu kwa 97.2%
- Muda wa majibu (wastani)
- 18.07s · haraka kwa 66.7%
Taarifa za modeli
Ilifanyiwa utafiti tarehe 2026-10-24
- Vigezo
- ~2.8T jumla (~104B vinavyotumika)
- Muundo
- MoE
- Upatikanaji
- Imefungwa
- Leseni
- -
Makadirio bora kulingana na ushahidi wa umma; mtoa huduma hakufichua thamani zote.
8.2
Uthabiti
8.4
8.5
Jumla ya tokeni za matokeo
114,612
Jumla ya tokeni za ingizo
249,927
Bei ya ingizo
$2.000 / 1M
Bei ya toleo
$10.000 / 1M
Bei ya kusoma akiba
$0.200 / 1M
Bei ya kuandika akiba
Haipo
Bei za akiba zinahusu tokeni za ingizo. Kusoma hutumia tena maelekezo yaliyohifadhiwa; kuandika huyahifadhi na kunaweza kuongeza gharama. Tokeni za matokeo hutumia bei ya matokeo.
Majaribio yasiyo thabiti
5
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.
#107 MoonshotAI: Kimi K3.1 Pro
high- Gharama
- $0.074
- Muda
- 141.9s
- Tokeni
- 7,498 tok
Historia ya uendeshaji
| Imepimwa tarehe | Alama | Uaminifu | Majaribio sahihi | Jumla ya gharama | Linganisha |
|---|---|---|---|---|---|
| 2026-10-24 20:16 Jaribio tena | 8.2 | 8.5 | $1.646 | Uendeshaji wa sasa | |
| 2026-10-24 15:44 Uendeshaji wa kwanza | 6.2 | 8.5 | $1.255 | 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) |
|---|---|---|---|---|---|---|---|---|---|---|
| 2026-10-24 20:16 · Uendeshaji wa sasa | marudio 69/69 | 8.2 | 8.4 | 8.5 | 16/23 | 5 | 114,612 | 249,927 | $1.646 | 54.31s |
| 2026-10-24 15:44 · Uendeshaji wa kwanza | marudio 69/69 | 6.2 | 5.3 | 8.5 | 7/23 | 13 | 92,327 | 165,632 | $1.255 | 39.12s |
| Tofauti | — | +2.0 | +3.1 | 0.0 | +9 | -8 | +22285 | +84295 | +$0.392 | +15190ms |
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
Kimi K3.1 ProhighvsSeed-2.0-LitemediumKimi K3.1 ProhighvsSeed-2.0-CodelowKimi K3.1 ProhighvsGPT-5.6 TerramediumKimi K3.1 ProhighvsDeepSeek V4 Pro 0423highKimi K3.1 ProhighvsGLM 5.2mediumKimi K3.1 ProhighvsClaude Opus 4.7mediumKimi K3.1 ProhighvsGrok 4.5highKimi K3.1 ProhighvsDeepSeek V4 Flash 0423highKimi K3.1 ProhighvsDeepSeek V4.1 Flash 1017maxKimi K3.1 ProhighvsDeepSeek V4.1 Flashlow
Mgawanyo wa kategoria
| Kategoria | Alama | Uthabiti | Majaribio sahihi |
|---|---|---|---|
| Kazi za mawakala | 7.4 | 3.8 | |
| Mbinu za kupinga AI | 10.0 | 10.0 | |
| Uandishi wa msimbo | 10.0 | 10.0 | |
| Mchanganyiko | 4.6 | 2.7 | |
| Uchanganuzi na uchimbaji wa data | 10.0 | 10.0 | |
| Mahususi kwa domeni | 5.3 | 7.2 | |
| Akili ya jumla | 10.0 | 10.0 | |
| Ufuataji wa maagizo | 10.0 | 10.0 | |
| Utatuzi wa mafumbo | 8.2 | 7.2 | |
| Mwito wa zana | 10.0 | 10.0 | |
| Maarifa ya jumla | 3.0 | 10.0 |