Yes — with 34.2 GB to spare
Gemma 3 27B at Q4_K_M fits your CPU only · DDR5 dual-channel entirely on the GPU at 8K context, at an estimated 2.4 tokens per second. There is room for its full 128K window.
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.
The VRAM budget
weights 15.4 GB
Weights 15.4 GB
KV cache @ 8K 1.03 GB
Runtime overhead 0.6 GB
Free 34.2 GB of 51.2 GB
Quantisation ladder
| Quant | Weights | Total @ 8K | Max context | Tok/s | Quality | Fit |
|---|---|---|---|---|---|---|
| Q8_0 | 27.1 GB | 28.7 GB | 128K | 1.4 | −0.1% ppl | Long context |
| Q6_K | 20.9 GB | 22.6 GB | 128K | 1.8 | −0.4% ppl | Long context |
| Q5_K_M | 18.1 GB | 19.7 GB | 128K | 2.1 | −0.8% ppl | Long context |
| Q4_K_M | 15.4 GB | 17.0 GB | 128K | 2.4 | −1.9% ppl | Recommended |
| Q3_K_M | 12.5 GB | 14.1 GB | 128K | 3.0 | −5.4% ppl | Long context |
| Q2_K | 10.7 GB | 12.3 GB | 128K | 3.5 | −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
$ llama-server \
-hf google/gemma-3-27b-it:Q4_K_M \
-c 8192 -ngl 99
The engine underneath most of the others. Every knob is exposed. More on llama.cpp.
01Download is 15.4 GB. Keep it on an SSD — a first load off a spinning disk takes minutes.
02With no GPU, thread count matters more than clock. Start at one thread per physical core.
03There is room to go to the model's full 128K context on this card.