Can the NVIDIA RTX 3080 10GB run Gemma 3 12B?

RTX 3080 10GB × Gemma 3 12B8K context
Yes, with heavy quantizationFits only at a low-quality quant
Memory at Q3_K_M9.8 GB of 10 GB
Weights 6 GBKV cache 3.2 GBRuntime 0.6 GB
QuantQ3_K_MNoticeably degraded
Memory9.8 GBof 10 GB usable
Speed~76tokens per second
  1. FP16
  2. Q8_0
  3. Q6_K
  4. Q5_K_M
  5. Q4_K_M
  6. Q3_K_M
  7. Q2_K

Only just. Gemma 3 12B fits on the NVIDIA RTX 3080 10GB at Q3_K_M, which costs noticeable quality. Expect about 76 tokens/second. A card with more memory is a better experience.

Run it with Ollama

Ollama's default gemma3:12b download is Q4_K_M, which needs about 11.2 GB, more than the 10 GB available. Download the Q3_K_M GGUF instead (LM Studio lists quants per model), or try a smaller context length.

Run it in the cloud

Want full quality? Run Gemma 3 12B on a bigger rented GPU.

Rent a cloud GPU by the hour. No upfront cost, stop any time.

Memory needed by quantization

At 8,192 tokens of context. The black line marks the 10 GB the NVIDIA RTX 3080 10GB can use.

QuantQualityMemoryFits?Speed
FP16Full quality28.2 GBToo big—
Q8_0Near-lossless16.8 GBToo big—
Q6_KExcellent13.8 GBToo big—
Q5_K_MVery good12.5 GBToo big—
Q4_K_MGood, most popular11.2 GBToo big—
Q3_K_MNoticeably degraded9.8 GBFits~76 tok/s
Q2_KHeavily degraded8.4 GBFits~100 tok/s

Frequently asked questions

Can NVIDIA RTX 3080 10GB run Gemma 3 12B?

Only just. Gemma 3 12B fits on the NVIDIA RTX 3080 10GB at Q3_K_M, which costs noticeable quality. Expect about 76 tokens/second. A card with more memory is a better experience.

How much VRAM does Gemma 3 12B need?

About 11.2 GB at Q4_K_M with 8,192 tokens of context, and about 28.2 GB at full FP16 precision.

How fast is Gemma 3 12B on the NVIDIA RTX 3080 10GB?

Roughly 76 tokens per second at Q3_K_M, estimated from the NVIDIA RTX 3080 10GB's 760 GB/s memory bandwidth.