Yes — with 5.4 GB to spare

Gemma 3 27B at Q4_K_M fits your A10 entirely on the GPU at 8K context, at an estimated 24 tokens per second. Past 76K 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 · 15.4 GB Gemma Terms of Use Released Mar 2025 Vision

The 2025 single-GPU generalist with vision. Its Gemma-licence terms are the reason to prefer Gemma 4 now.

What hardware do I need for Gemma 3 27B? →

The VRAM budget

weights 15.4 GB

Weights 15.4 GB KV cache @ 8K 1.03 GB Runtime overhead 0.6 GB Free 5.4 GB of 22.4 GB

Quantisation ladder

Quant	Weights	Total @ 8K	Max context	Tok/s	Quality	Fit
Q8_0	27.1 GB	28.7 GB	—	~4.3	−0.1% ppl	6.3 GB over
Q6_K	20.9 GB	22.6 GB	6K	~16	−0.4% ppl	0.2 GB over
Q5_K_M	18.1 GB	19.7 GB	42K	20	−0.8% ppl	Long context
Q4_K_M	15.4 GB	17.0 GB	76K	24	−1.9% ppl	Recommended
Q3_K_M	12.5 GB	14.1 GB	114K	29	−5.4% ppl	Long context
Q2_K	10.7 GB	12.3 GB	128K	34	−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. This model interleaves sliding-window layers (1024 tokens, 1 global in 6), which is why its cache barely grows with context.

How to run it

Ollama llama.cpp LM Studio vLLM

terminal

$ ollama pull gemma3:27b
$ OLLAMA_CONTEXT_LENGTH=8192 \
    ollama run gemma3:27b

The default. One binary, a model registry, an OpenAI-compatible port. More on Ollama.

01Download is 15.4 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.

03There is room to go to 76K context on this card.

See all models for this rig Compare with another model