LVBench
Ergebnisse von KI-Benchmarks und Modellleistung
Typ: percentageLong Context
Erfasste Ergebnisse
41
Durchschnittspunktzahl
58.40%
Punktzahlbereich
45.30% - 91.70%
Führendes Modell
91.70% - Gemini 4 Argon
Punktzahlen im Zeitverlauf
Einzelne Benchmark-Punktzahlen nach Datum.
Modellergebnisse
| Organisation | Modelle | Gemeldet am | Höchste Punktzahl | Info | Selbst gemeldet | Quelle |
|---|---|---|---|---|---|---|
| Gemini 4 Argon | 13. Okt. 2026 | 91,70 % | Google-reported launch evaluation; source chart: https://storage.googleapis.com/gweb-uniblog-publish-prod/original_images/gemini-4-argon_table_blog.gif | Ja | Quelle | |
| Seed 2.0 Pro | 14. Feb. 2026 | 76,40 % | Seed2 official benchmark table | LVBench | Ja | Quelle | |
| Kimi K2.5 | 27. Jan. 2026 | 75,90 % | - | Ja | Quelle | |
| Qwen 3.5 122B A10B | 24. Feb. 2026 | 74,40 % | - | Ja | Quelle | |
| Qwen 3.5 27B | 24. Feb. 2026 | 73,60 % | - | Ja | Quelle | |
| Qwen 3.5 Flash | 23. Feb. 2026 | 73,60 % | inferred family alias from qwen3.5-27b (score=0.4147; benches=81) | Ja | Quelle | |
| Seed 2.0 Lite | 14. Feb. 2026 | 73 % | Seed2 official benchmark table | LVBench | Ja | Quelle | |
| Qwen 3.5 35B A3B | 24. Feb. 2026 | 71,40 % | - | Ja | Quelle | |
| Qwen 3 VL 235B A22B Instruct | - | 67,70 % | - | Ja | - | |
| Seed 2.0 Mini | 14. Feb. 2026 | 66,60 % | Seed2 official benchmark table | LVBench | Ja | Quelle | |
| Qwen 3 VL 32B Instruct | - | 63,80 % | - | Ja | - | |
| Qwen 3 VL 235B A22B Thinking | - | 63,60 % | - | Ja | - | |
| Qwen 3 VL 32B Thinking | - | 62,60 % | - | Ja | - | |
| Qwen 3 VL 30B A3B Instruct | - | 62,50 % | - | Ja | - | |
| Qwen 3 VL 30B A3B Thinking | - | 59,20 % | - | Ja | - | |
| Qwen 3 VL 8B Instruct | - | 58 % | - | Ja | - | |
| Qwen 3 VL 4B Instruct | - | 56,20 % | - | Ja | - | |
| Qwen 3 8B | - | 55,80 % | inferred high-confidence family alias from qwen3-vl-8b-thinking (score=0.4600; benches=50) | Ja | - | |
| Qwen 3 Guard Stream 8B | - | 55,80 % | inferred family alias from qwen3-vl-8b-thinking (score=0.3371; benches=50) | Ja | - | |
| Qwen 3 VL 8B Thinking | - | 55,80 % | - | Ja | - | |
| Qwen 3 VL Embedding 8B | - | 55,80 % | inferred high-confidence family alias from qwen3-vl-8b-thinking (score=0.5232; benches=50) | Ja | - | |
| Qwen 3 VL Reranker 8B | - | 55,80 % | inferred high-confidence family alias from qwen3-vl-8b-thinking (score=0.5275; benches=50) | Ja | - | |
| Qwen 3 4B Instruct 2507 | - | 53,50 % | inferred family alias from qwen3-vl-4b-thinking (score=0.3462; benches=48) | Ja | - | |
| Qwen 3 4B SafeRL | - | 53,50 % | inferred family alias from qwen3-vl-4b-thinking (score=0.3850; benches=48) | Ja | - | |
| Qwen 3 4B Thinking 2507 | - | 53,50 % | inferred family alias from qwen3-vl-4b-thinking (score=0.3462; benches=48) | Ja | - |
