Gemma 3 12B VRAM and GPU Requirements

Gemma 3 12B needs about 8.20 GB at Q4_K_M with 8K context.

Which GPUs fit

VRAM by quantization

Totals at 8,192 tokens. Q4_K_M is the usual balance of size and quality. FP16 weights are the parameter count times 2 bytes. The FP16 column in the table is the full estimate: weights, cache, and 1 GB.

Context length

Q4_K_M total as the context changes. Cache is the part that grows.

• 2,048 7.82 GB
• 8,192 8.20 GB
• 32,768 9.70 GB
• 131,072 15.70 GB

Gemma 3 12B hardware notes

• Unified memory counts 75 percent of the listed pool as usable. Discrete cards use the full listed memory. The memory column still shows the listed figure.
• Last updated 2025-03-21.
• Weights, cache, and the 1 GB overhead are estimates. How this is calculated.
• Architecture numbers come from unsloth/gemma-3-12b-it because the official download is gated. The parameter count is the official Hugging Face record.

Gemma 3 12B VRAM FAQ

How much VRAM does Gemma 3 12B need at Q4_K_M?

8.20 GB at 8,192 tokens. Weights are 6.38 GB, the cache is 0.81 GB, and overhead is 1 GB.

Does a 24 GB GPU fit Gemma 3 12B at Q4_K_M?

GeForce RTX 3090 24GB is Fits, 15.80 GB left. The quant table uses 8,192 tokens, the smaller of 8,192 and the configured maximum of 131,072.

What context length is configured for Gemma 3 12B?

The configured context length is 131,072 tokens. The quant table uses 8,192 tokens, the smaller of 8,192 and the configured maximum of 131,072.

Which source supplies the parameter count for Gemma 3 12B?

12.19B parameters, from the Hugging Face model record for google/gemma-3-12b-it. Architecture numbers come from the public copy at unsloth/gemma-3-12b-it, because the official download is gated. The parameter count is the official record.

Related models