RUN THIS LLM
Search local LLM hardware requirements
CodeLlama 13B
Meta · 13B · Code
Mid-size code model with stronger reasoning than 7B. Good for complex code generation tasks.
VRAM Requirements
| Quantization | VRAM |
|---|---|
| Q4_K_M (smallest) | 7.8 GB |
| Q8_0 (balanced) | 14.3 GB |
| FP16 (full quality) | 26 GB |
Specifications
- Parameters: 13B
- Category: Code
- Max context: 16K tokens
- System RAM: 16 GB minimum
- HuggingFace: codellama/CodeLlama-13b-Instruct-hf
Benchmarks
- HumanEval: 42 — Code generation (pass@1)
Loading interactive analysis...