Benchmark
Gemma 3 270M
- Google
- Veröffentlichung · 14. August 2025
- Offene Gewichte
- Ohne Reasoning
Intelligenz
Intelligence Index
#254 von 266- 5,25,2
LFM2.5-1.2B-Instruct - 5,25,2
Jamba 1.7 Mini - 5,25,2
Olmo 3 7B Instruct - 5,25,2
Granite 4.0 H 1B - 5,25,2
Exaone 4.0 1.2B (Non-reasoning) - 5,15,1
Gemma 3 270M - 5,15,1
DeepHermes 3 - Llama-3.1 8B Preview (Non-reasoning) - 5,15,1Apertus 70B Instruct
- 5,15,1
Granite 4.0 Micro - 55
Molmo2-8B
Daten von Artificial Analysis- AA-Omniscience Accuracy0,9%#258
- Humanity's Last Exam3,6%#256
- AA-LCR0%#229
- CritPt0%#143
- GPQA Diamond22,4%#213
- IFBench12,1%#153
- τ²-Bench Telecom9,1%#139
- Terminal-Bench Hard0%#121
Gemma 3 270MMedian
Daten von Artificial AnalysisÄhnliche Modelle