HumanEval 92% — the best single-GPU coding model for RTX 5090 owners. Excels at multi-file edits, refactoring, and agentic tool use pipelines. The definitive open coding model at the 20GB VRAM tier.
This model requires a specialized High-VRAM environment. Ensure you have the latest CUDA Drivers or Metal Framework installed.
Minimum VRAM: 22GB VRAM Recommended
The Qwen 2.5 Coder 32B model by Alibaba is a 32B dense parameter architecture optimized for code tasks. It requires approximately 20GB of VRAM to comfortably run locally using a Q4_K_M quantization. Extending the context window up to 131,072 tokens will dynamically allocate further VRAM, meaning high-bandwidth memory hardware is strictly advised.
For running Qwen 2.5 Coder 32B at maximum tokens-per-second, we recommend using LM Studio or Ollama with a GGUF quantization (Q4_K_M or Q6_K). If you are multi-GPU, use vLLM to distribute the layers across your VRAM pool for optimal throughput.
You will need a GPU with at least 22GB of VRAM to run the Q4_K_M quantized version smoothly with a moderate context window.
The simplest method is utilizing Ollama by executing 'ollama run qwen2.5-coder:32b' directly in your command line. Alternatively, you can search for the model via LM Studio's interface.