Qwen2.5-Coder 32B Instruct

Running Qwen2.5-Coder 32B Instruct locally

At Q4_K_M and 4K of context, Qwen2.5-Coder 32B Instruct needs about 21.7 GB — 18.7 GB of weights, 1.00 GB of KV cache and a 2.0 GB activation buffer. The smallest card in this index that clears that comfortably is the RTX 5090 at 32 GB, and 24 of the 63 cards here do. These are calculated figures, not measurements: the estimate stops counting a card as comfortable at 88% of its VRAM, which is roughly the room a desktop session needs.

What longer context costs

Going from 4K to 32K adds about 7.00 GB, taking the total to 29.4 GB. Weights do not move with context — only the KV cache does, and it grows linearly, so this is the number to watch when planning for long documents. The model's native window is 128K; holding all of it at Q4_K_M would need about 56 GB.

Which build to download

This index tracks 3 formats for it — GGUF, EXL2, AWQ — across 3 levels. Q4_K_M carries the lowest published perplexity loss at 2.5%. The fastest level measured here is EXL2 3.5bpw at 65 tok/s on an RTX 4090, batch 1. GGUF runs on llama.cpp and Ollama across NVIDIA, AMD and Apple silicon; AWQ and GPTQ target vLLM on CUDA and ROCm; EXL2 is ExLlamaV2 and CUDA only.

Common questions

How much VRAM does Qwen2.5-Coder 32B Instruct need?

About 21.7 GB at Q4_K_M with 4K of context and batch 1, rising to roughly 29.4 GB at 32K. That figure is weights plus KV cache plus a 10% activation buffer, calculated from the model's architecture rather than measured on a card.

Will Qwen2.5-Coder 32B Instruct run on a 32GB GPU?

Yes — at Q4_K_M and 4K context it needs about 21.7 GB, which leaves 10.3 GB spare on a RTX 5090. That is the smallest card in this index that clears it comfortably; 24 of 63 do.

Which quantization of Qwen2.5-Coder 32B Instruct should I use?

Q4_K_M has the lowest published quality loss (2.5%), and Q4_K_M is the level most people run. All 3 levels in the index are Q4_K_M, EXL2 3.5bpw, AWQ INT4.