RUN THIS LLM
Search local LLM hardware requirements
Gemma 2 9B
Google · 9B · General
Previous-gen 9B model. Still very competitive for general chat and instruction following.
VRAM Requirements
| Quantization | VRAM |
|---|---|
| Q4_K_M (smallest) | 5.4 GB |
| Q8_0 (balanced) | 9.9 GB |
| FP16 (full quality) | 18 GB |
Specifications
- Parameters: 9B
- Category: General
- Max context: 8K tokens
- System RAM: 12 GB minimum
- HuggingFace: google/gemma-2-9b-it
Benchmarks
- MMLU: 72 — General knowledge & reasoning (5-shot)
- HumanEval: 54 — Code generation (pass@1)
- MATH: 36 — Competition-level math reasoning
Loading interactive analysis...