Alibaba-Modelle: Intelligenz, Leistung und Preis
Qwen3.8 Max (0902)Qwen3.8 27B (xhigh)Qwen3.8 27B (medium)Qwen3.8 27B (low)Qwen3.8 27B (non-reasoning)Artificial Analysis Intelligence Index
Qwen3.8 27B (non-reasoning)Qwen3.8 27B (low)Qwen3.8 27B (medium)Qwen3.8 27B (xhigh)Qwen3.8 Max (0902)Pro Aufgabe des Intelligence Index
Qwen3.8 27B (low)Qwen3.8 27B (medium)Qwen3.8 27B (xhigh)Qwen3.8 27B (non-reasoning)Qwen3.8 Max (0902)Pro Aufgabe des Intelligence Index
Artificial Analysis hat 89 Modelle von Alibaba gebenchmarkt. Unten werden die wichtigsten Kennzahlen dieser Modelle verglichen.
- Bei der Intelligenz ist Qwen3.8 Max (0902) mit 34 das führende Modell von Alibaba.
- Bei der Geschwindigkeit hat Qwen3.8 27B (Non-reasoning) mit 18,3 Min. die niedrigste Zeit pro Aufgabe.
- Bei der Latenz bietet Qwen3.5 9B (Non-reasoning) mit 1,03 s die kürzeste Zeit bis zum ersten Antworttoken.
- Beim Preis bietet Qwen3.8 27B (Low) mit $1.91 die niedrigsten Kosten pro Aufgabe. Die Preise unterscheiden sich zwischen den Modellen um einen Faktor von bis zu 5,3.
Intelligenz
Artificial Analysis Intelligence Index
Intelligence Index vs. Kosten pro Aufgabe
Kosten
Kosten pro Aufgabe des Intelligence Index
Geschwindigkeit und Latenz
Zeit pro Aufgabe des Intelligence Index
Fähigkeitswerte
Fähigkeitsindizes
Incorporates 5 evaluations: AA-Briefcase v1.1, AA-Omniscience v1.1, HLE-Diamond, AutomationBench-AA, GDP.pdf · Higher is better
Incorporates 4 evaluations: AutomationBench-AA, AA-Briefcase v1.1, AA-Omniscience v1.1, GDP.pdf · Higher is better
Incorporates 5 evaluations: AA-Omniscience v1.1, AA-Briefcase v1.1, HLE-Diamond, GDP.pdf, AutomationBench-AA · Higher is better
Incorporates 5 evaluations: AA-Omniscience v1.1, AA-Briefcase v1.1, MLCR-AA, AutomationBench-AA, HLE-Diamond · Higher is better
Incorporates 7 evaluations: AA-Omniscience v1.1, CritPt v2 (Hard Subset), Terminal-Bench-Science 0.1, HLE-Diamond, Terminal-Bench 4.0, FrontierCode 2.0, AA-Briefcase v1.1 · Higher is better
Incorporates 4 evaluations: AA-Omniscience v1.1, AA-Briefcase v1.1, HLE-Diamond, GDP.pdf · Higher is better
Alle Veröffentlichungen von Alibaba
Qwen3.8 Max (0902)
1 Modell
34
Höchste Intelligenz
Qwen3.8-Flash-Next
1 Modell
Qwen3.8 27B
4 Modelle
20
Höchste Intelligenz
Qwen3.8 2.4T A95B
1 Modell
Qwen3.8 Max
1 Modell
Qwen3.7 Plus
1 Modell
Qwen3.7 Max
1 Modell
Qwen3.6 27B
2 Modelle
Qwen3.6 Max Preview
1 Modell
Qwen3.6 35B A3B
2 Modelle
Qwen3.6 Plus
1 Modell
Qwen3.5 Omni Flash
1 Modell
Qwen3.5 Omni Plus
1 Modell
Qwen3.5 0.8B
2 Modelle
Qwen3.5 2B
2 Modelle
Qwen3.5 4B
2 Modelle
Qwen3.5 9B
2 Modelle
Qwen3.5 122B A10B
2 Modelle
Qwen3.5 27B
2 Modelle
Qwen3.5 35B A3B
2 Modelle
Qwen3.5 397B A17B
2 Modelle
Qwen3 Coder Next
1 Modell
Qwen3 Max Thinking
1 Modell
Qwen3 Max Thinking (Preview)
1 Modell
Qwen3 VL 32B (Reasoning)
1 Modell
Qwen3 VL 32B Instruct
1 Modell
Qwen3 VL 4B (Reasoning)
1 Modell
Qwen3 VL 4B Instruct
1 Modell
Qwen3 VL 8B (Reasoning)
1 Modell
Qwen3 VL 8B Instruct
1 Modell
Qwen3 VL 30B A3B (Reasoning)
1 Modell
Qwen3 VL 30B A3B Instruct
1 Modell
Qwen3 Max
1 Modell
Qwen3 VL 235B A22B (Reasoning)
1 Modell
Qwen3 VL 235B A22B Instruct
1 Modell
Qwen3 Omni 30B A3B (Reasoning)
1 Modell
Qwen3 Omni 30B A3B Instruct
1 Modell
Qwen3 Next 80B A3B (Reasoning)
1 Modell
Qwen3 Next 80B A3B Instruct
1 Modell
Qwen3 Max (Preview)
1 Modell
Qwen3 4B 2507 Instruct
1 Modell
Qwen3 4B 2507 (Reasoning)
1 Modell
Qwen3 Coder 30B A3B Instruct
1 Modell
Qwen3 30B A3B 2507 (Reasoning)
1 Modell
Qwen3 30B A3B 2507 Instruct
1 Modell
Qwen3 235B A22B 2507 (Reasoning)
1 Modell
Qwen3 Coder 480B A35B Instruct
1 Modell
Qwen3 235B A22B 2507 Instruct
1 Modell
Qwen3 0.6B
2 Modelle
Qwen3 1.7B
2 Modelle
Qwen3 14B
2 Modelle
Qwen3 235B A22B
2 Modelle
Qwen3 30B A3B
2 Modelle
Qwen3 32B
2 Modelle
Qwen3 4B
2 Modelle
Qwen3 8B
2 Modelle
QwQ 32B
1 Modell
Qwen2.5 Max
1 Modell
QwQ 32B-Preview
1 Modell
Qwen2.5 Turbo
1 Modell
Qwen2.5 Coder Instruct 32B
1 Modell
Qwen2.5 Instruct 72B
1 Modell
Qwen2.5 Coder Instruct 7B
1 Modell
Qwen2.5 Instruct 32B
1 Modell
Qwen2 Instruct 72B
1 Modell
Qwen1.5 Chat 110B
1 Modell
Qwen Chat 72B
1 Modell
Qwen Chat 14B
1 Modell
Weitere Details
Gewichte | Anbieter-Benchmarks | |||||||||
|---|---|---|---|---|---|---|---|---|---|---|
| Qwen3.8 Max (0902) | Alibaba | 34 | - | 984k | 1,2 $ | 61,1 | 47 | Nicht verfügbar | | |
| Qwen3.8 27B (Xhigh) | Alibaba | 20 | 27B | 256k | 0,5 $ | 42,8 | 43 |   +7 | ||
| Qwen3.8 27B (Medium) | Alibaba | 17 | 27B | 256k | 0,5 $ | 33,3 | 43 | | ||
| Qwen3.8 27B (Low) | Alibaba | 16 | 27B | 256k | 0,5 $ | 28,9 | 46 | | ||
| Qwen3.8 27B (Non-reasoning) | Alibaba | 10 | 27B | 256k | 0,5 $ | 18,3 | 41 |   +4 | ||
| Qwen3.5 Omni Plus | Alibaba | - | - | 256k | 0,8 $ | - | 82 | Nicht verfügbar | | |
| Qwen3.5 122B A10B (Non-reasoning) | Alibaba | - | 125B 10B während der Inferenz aktiv | 262k | 0,7 $ | - | 105 | | ||
| Qwen2.5 Coder Instruct 32B | Alibaba | - | 32B | 131k | - | - | - | - | ||
| Qwen3 Coder 480B A35B Instruct | Alibaba | - | 480B 35B während der Inferenz aktiv | 262k | 0,4 $ | - | 69 |  | ||
| Qwen3.5 2B (Non-reasoning) | Alibaba | - | 2.3B | 262k | - | - | - | - | ||
| Qwen3 Omni 30B A3B Instruct | Alibaba | - | 35.3B 3B während der Inferenz aktiv | 66k | 0,3 $ | - | 71 | | ||
| Qwen3 0.6B (Non-reasoning) | Alibaba | - | 0.8B | 32k | - | - | - | - | ||
| Qwen3.5 397B A17B (Reasoning) | Alibaba | - | 397B 17B während der Inferenz aktiv | 262k | 0,9 $ | - | 52 |  +2 | ||
| Qwen3.5 27B (Reasoning) | Alibaba | - | 27.8B | 262k | 0,5 $ | - | 63 |   +2 | ||
| Qwen3 4B (Reasoning) | Alibaba | - | 4B | 32k | - | - | - | - | ||
| Qwen2.5 Instruct 32B | Alibaba | - | 32B | 128k | - | - | - | - | ||
| Qwen3.5 9B (Non-reasoning) | Alibaba | - | 9.7B | 262k | 0,2 $ | - | 125 |  | ||
| Qwen3.5 122B A10B (Reasoning) | Alibaba | - | 125B 10B während der Inferenz aktiv | 262k | 0,7 $ | - | 109 |  | ||
| Qwen3 Next 80B A3B Instruct | Alibaba | - | 80B 3B während der Inferenz aktiv | 262k | 0,2 $ | - | 110 |  | ||
| Qwen3 4B 2507 (Reasoning) | Alibaba | - | 4B | 262k | - | - | - | - | ||
| Qwen3 1.7B (Non-reasoning) | Alibaba | - | 2B | 32k | - | - | - | - | ||
| Qwen3.5 9B (Reasoning) | Alibaba | - | 9.7B | 262k | 0,1 $ | - | 88 | | ||
| Qwen3.5 4B (Non-reasoning) | Alibaba | - | 4.7B | 262k | 0,0 $ | - | 28 | | ||
| Qwen3.5 397B A17B (Non-reasoning) | Alibaba | - | 397B 17B während der Inferenz aktiv | 262k | 0,9 $ | - | 47 |  | ||
| Qwen3 235B A22B 2507 Instruct | Alibaba | - | 235B 22B während der Inferenz aktiv | 256k | 0,3 $ | - | 35 |  +2 | ||
| Qwen2.5 Turbo | Alibaba | - | - | 1M | - | - | - | Nicht verfügbar | - | |
| Qwen3.6 27B (Non-reasoning) | Alibaba | - | 27.8B | 262k | 0,9 $ | - | 45 | | ||
| Qwen3 VL 8B Instruct | Alibaba | - | 8.8B | 256k | - | - | - | - | ||
| Qwen3 30B A3B (Reasoning) | Alibaba | - | 30.5B 3.3B während der Inferenz aktiv | 33k | 0,2 $ | - | 87 |  | ||
| Qwen3.5 0.8B (Non-reasoning) | Alibaba | - | 0.9B | 262k | - | - | - | - | ||
| Qwen Chat 72B | Alibaba | - | 72B | 34k | - | - | - | - | ||
| Qwen3 235B A22B (Non-reasoning) | Alibaba | - | 235B 22B während der Inferenz aktiv | 33k | - | - | - | - | ||
| Qwen3 14B (Reasoning) | Alibaba | - | 14.8B | 33k | 0,1 $ | - | 52 | | ||
| Qwen3.5 35B A3B (Reasoning) | Alibaba | - | 36B 3B während der Inferenz aktiv | 262k | 0,4 $ | - | 123 | +2 | ||
| Qwen3 0.6B (Reasoning) | Alibaba | - | 0.8B | 32k | - | - | - | - | ||
| Qwen3.5 27B (Non-reasoning) | Alibaba | - | 27.8B | 262k | 0,5 $ | - | 53 | | ||
| Qwen3 VL 30B A3B Instruct | Alibaba | - | 30B 3B während der Inferenz aktiv | 256k | 0,4 $ | - | 37 | | ||
| Qwen3 1.7B (Reasoning) | Alibaba | - | 2B | 32k | - | - | - | - | ||
| Qwen3 Max (Preview) | Alibaba | - | - | 262k | - | - | - | Nicht verfügbar | - | |
| Qwen3.5 2B (Reasoning) | Alibaba | - | 2.3B | 262k | - | - | - | - | ||
| Qwen3 30B A3B 2507 (Reasoning) | Alibaba | - | 30.5B 3.3B während der Inferenz aktiv | 262k | - | - | - | - | ||
| Qwen3.8 Max | Alibaba | - | - | 1M | 1,2 $ | - | 49 | Nicht verfügbar | | |
| Qwen3.6 Max Preview | Alibaba | - | - | 256k | - | - | - | Nicht verfügbar | - | |
| Qwen3.5 4B (Reasoning) | Alibaba | - | 4.7B | 262k | 0,0 $ | - | 33 | | ||
| Qwen3.6 35B A3B (Non-reasoning) | Alibaba | - | 36B 3B während der Inferenz aktiv | 262k | 0,6 $ | - | 95 |  +4 | ||
| Qwen3.7 Max | Alibaba | - | - | 1M | 1,6 $ | - | 162 | Nicht verfügbar | | |
| Qwen3 VL 32B Instruct | Alibaba | - | 33.4B | 256k | - | - | - | - | ||
| Qwen3.5 Omni Flash | Alibaba | - | - | 256k | 0,2 $ | - | 144 | Nicht verfügbar | | |
| Qwen2 Instruct 72B | Alibaba | - | 72B | 131k | - | - | - | - | ||
| Qwen3 Max | Alibaba | - | - | 262k | - | - | - | Nicht verfügbar | - | |
| Qwen3 30B A3B 2507 Instruct | Alibaba | - | 30.5B 3.3B während der Inferenz aktiv | 262k | 0,1 $ | - | 76 |  | ||
| Qwen3 Max Thinking | Alibaba | - | - | 256k | - | - | - | Nicht verfügbar | - | |
| Qwen3.6 35B A3B (Reasoning) | Alibaba | - | 36B 3B während der Inferenz aktiv | 262k | 0,6 $ | - | 111 | +5 | ||
| QwQ 32B-Preview | Alibaba | - | 32.8B | 33k | - | - | - | - | ||
| Qwen3.6 27B (Reasoning) | Alibaba | - | 27.8B | 262k | 0,9 $ | - | 51 | | ||
| Qwen3.8 2.4T A95B | Alibaba | - | 2.4T 95B während der Inferenz aktiv | 262k | 1,2 $ | - | 40 |  +2 | ||
| Qwen3.8-Flash-Next | Alibaba | - | 180B 6B während der Inferenz aktiv | 256k | 0,1 $ | - | 86 | | ||
| Qwen3 8B (Reasoning) | Alibaba | - | 8.2B | 131k | - | - | - | - | ||
| Qwen3.5 0.8B (Reasoning) | Alibaba | - | 0.9B | 262k | - | - | - | - | ||
| Qwen3 VL 235B A22B (Reasoning) | Alibaba | - | 235B 22B während der Inferenz aktiv | 262k | - | - | - | - | ||
| Qwen3.7 Plus | Alibaba | - | - | 1M | 0,3 $ | - | 54 | Nicht verfügbar | | |
| Qwen3 32B (Reasoning) | Alibaba | - | 32.8B | 33k | 0,3 $ | - | 23 |  | ||
| Qwen3 4B 2507 Instruct | Alibaba | - | 4B | 262k | - | - | - | - | ||
| Qwen3 8B (Non-reasoning) | Alibaba | - | 8.2B | 131k | 0,2 $ | - | - | | ||
| Qwen2.5 Instruct 72B | Alibaba | - | 72B | 131k | 0,5 $ | - | 22 | | ||
| Qwen3 Omni 30B A3B (Reasoning) | Alibaba | - | 35.3B 3B während der Inferenz aktiv | 66k | 0,3 $ | - | 86 | | ||
| Qwen2.5 Max | Alibaba | - | - | 32k | - | - | - | Nicht verfügbar | - | |
| Qwen3 235B A22B (Reasoning) | Alibaba | - | 235B 22B während der Inferenz aktiv | 33k | - | - | - | - | ||
| Qwen3 32B (Non-reasoning) | Alibaba | - | 32.8B | 33k | 0,2 $ | - | 94 | | ||
| Qwen Chat 14B | Alibaba | - | 14B | 8k | - | - | - | - | ||
| Qwen3 Next 80B A3B (Reasoning) | Alibaba | - | 80B 3B während der Inferenz aktiv | 262k | - | - | - | - | ||
| Qwen3 Max Thinking (Preview) | Alibaba | - | - | 262k | - | - | - | Nicht verfügbar | - | |
| Qwen3 VL 30B A3B (Reasoning) | Alibaba | - | 30B 3B während der Inferenz aktiv | 256k | 0,5 $ | - | - | | ||
| QwQ 32B | Alibaba | - | 32.8B | 131k | - | - | - | - | ||
| Qwen1.5 Chat 110B | Alibaba | - | 110B | 32k | - | - | - | - | ||
| Qwen3 VL 32B (Reasoning) | Alibaba | - | 33.4B | 256k | - | - | - | - | ||
| Qwen3 VL 235B A22B Instruct | Alibaba | - | 235B 22B während der Inferenz aktiv | 262k | 0,3 $ | - | 38 | | ||
| Qwen3 Coder 30B A3B Instruct | Alibaba | - | 30.5B 3.3B während der Inferenz aktiv | 262k | 0,2 $ | - | 169 | | ||
| Qwen3 VL 8B (Reasoning) | Alibaba | - | 8.8B | 256k | - | - | - | - | ||
| Qwen2.5 Coder Instruct 7B | Alibaba | - | 7.6B | 131k | - | - | - | - | ||
| Qwen3 4B (Non-reasoning) | Alibaba | - | 4B | 32k | - | - | - | - | ||
| Qwen3 14B (Non-reasoning) | Alibaba | - | 14.8B | 33k | 0,1 $ | - | 45 | | ||
| Qwen3.5 35B A3B (Non-reasoning) | Alibaba | - | 36B 3B während der Inferenz aktiv | 262k | 0,4 $ | - | 101 | | ||
| Qwen3 30B A3B (Non-reasoning) | Alibaba | - | 30.5B 3.3B während der Inferenz aktiv | 33k | 0,2 $ | - | 86 | | ||
| Qwen3.6 Plus | Alibaba | - | - | 1M | 0,4 $ | - | 52 | Nicht verfügbar | | |
| Qwen3 VL 4B Instruct | Alibaba | - | 4.4B | 256k | - | - | - | - | ||
| Qwen3 235B A22B 2507 (Reasoning) | Alibaba | - | 235B 22B während der Inferenz aktiv | 256k | 0,4 $ | - | 133 | | ||
| Qwen3 VL 4B (Reasoning) | Alibaba | - | 4.4B | 256k | - | - | - | - | ||
| Qwen3 Coder Next | Alibaba | - | 79.7B 3B während der Inferenz aktiv | 256k | 0,4 $ | - | 70 | |