Yes — with 6.2 GB to spare

Gemma 4 12B at Q4_K_M fits your Radeon RX 7900 GRE entirely on the GPU at 8K context, at an estimated 52 tokens per second. Past 107K 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 · 6.7 GB Apache 2.0 Released 29 May 2026 Vision

The "unified" Gemma 4: text, image and audio in one 12B that fits a 12 GB card at Q4. 140+ languages.

What hardware do I need for Gemma 4 12B? →

The VRAM budget

weights 6.7 GB

Weights 6.7 GB KV cache @ 8K 0.81 GB Runtime overhead 0.6 GB Free 6.2 GB of 14.4 GB

Quantisation ladder

Quant	Weights	Total @ 8K	Max context	Tok/s	Quality	Fit
F16	22.4 GB	23.8 GB	—	~3.4	Reference	9.4 GB over
Q8_0	11.9 GB	13.3 GB	25K	29	−0.1% ppl	Long context
Q6_K	9.2 GB	10.6 GB	69K	38	−0.4% ppl	Long context
Q5_K_M	7.9 GB	9.3 GB	89K	44	−0.8% ppl	Long context
Q4_K_M	6.7 GB	8.2 GB	107K	52	−1.9% ppl	Recommended
Q3_K_M	5.5 GB	6.9 GB	128K	64	−5.4% ppl	Long context
Q2_K	4.7 GB	6.1 GB	140K	75	−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, 8 of 48 layers global), which is why its cache barely grows with context.

How to run it

Ollama llama.cpp LM Studio vLLM

terminal

$ ollama pull gemma4:12b
$ OLLAMA_CONTEXT_LENGTH=8192 \
    ollama run gemma4:12b

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

01Download is 6.7 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 107K context on this card.

See all models for this rig Compare with another model