Gemini 1.5 Flash 8B
A multimodal model capable of processing audio, images, video, and text with high efficiency. Features JSON mode, function calling, code execution, and system instructions support. Optimized for fast inference with 8B parameters.
Benchmark results
| Benchmark | Score | Tags | Source |
|---|---|---|---|
| FLEURS | 86.4% | self-reported llm-stats | link → |
| GPQA | 38.4% | self-reported llm-stats | link → |
| HiddenMath | 32.8% | self-reported llm-stats | link → |
| MATH | 58.7% | self-reported llm-stats | link → |
| MathVista | 54.7% | self-reported llm-stats | link → |
| MMLU-Pro | 58.7% | self-reported llm-stats | link → |
| MMMU | 53.7% | self-reported llm-stats | link → |
| MRCR | 54.7% | self-reported llm-stats | link → |
| Natural2Code | 75.5% | self-reported llm-stats | link → |
| Vibe-Eval | 40.9% | self-reported llm-stats | link → |
| Video-MME | 66.2% | self-reported llm-stats | link → |
| WMT23 | 72.6% | self-reported llm-stats | link → |
| XSTest | 92.6% | self-reported llm-stats | link → |