Yes, just — 1.4 GB spare

Qwen2.5-Coder 32B at Q4_K_M fits your Tesla P40 entirely on the GPU at 8K context, at an estimated 11 tokens per second. Past 13K the KV cache pushes it over — quantise the cache to q8_0, or step down a quantisation, to go longer.

Fully on GPU 8K context Q4_K_M · 18.4 GB Apache 2.0 Released Nov 2024

The first local code model that felt competitive with hosted assistants. Qwen3.8 27B is smaller and far ahead on agentic work.

What hardware do I need for Qwen2.5-Coder 32B? →

The VRAM budget

weights 18.4 GB

Weights 18.4 GB KV cache @ 8K 2.00 GB Runtime overhead 0.6 GB Free 1.4 GB of 22.4 GB

Quantisation ladder

Quant	Weights	Total @ 8K	Max context	Tok/s	Quality	Fit
Q8_0	32.5 GB	35.1 GB	—	~2.3	−0.1% ppl	12.7 GB over
Q6_K	25.0 GB	27.6 GB	—	~4.2	−0.4% ppl	5.2 GB over
Q5_K_M	21.7 GB	24.3 GB	—	~6.9	−0.8% ppl	1.9 GB over
Q4_K_M	18.4 GB	21.0 GB	13K	11	−1.9% ppl	Recommended
Q3_K_M	14.9 GB	17.5 GB	27K	14	−5.4% ppl	Long context
Q2_K	12.8 GB	15.4 GB	36K	16	−15% ppl	Long context

Quality is the published perplexity delta against f16 weights. Max context assumes an f16 KV cache; q8_0 roughly doubles it.

How to run it

Ollama llama.cpp LM Studio vLLM

terminal

$ llama-server \
    -hf Qwen/Qwen2.5-Coder-32B-Instruct:Q4_K_M \
    -c 8192 -ngl 99

The engine underneath most of the others. Every knob is exposed. More on llama.cpp.

01Download is 18.4 GB. Keep it on an SSD — a first load off a spinning disk takes minutes.

02Close anything else holding VRAM. A browser with hardware acceleration can sit on 1–2 GB.

03Only 1.4 GB is spare, so a long system prompt can still push it over. Its real ceiling here is 13K context.

See all models for this rig Compare with another model