Gemma on NVIDIA Blackwell

Can I run Gemma 3 4B on an RTX 5080?

Yes. Gemma 3 4B at Q4_K_M uses 3.6 GB of the 14.2 GB available on a RTX 5080, and runs at about 226 tokens per second. You can push the context to 128K.

Runs comfortably

3.6 GB of 14.2 GB · 25%
014 GB
Weights 2.5 GB
KV cache 0.3 GB
Runtime overhead 0.8 GB

Gemma 3 4B at Q4_K_M leaves 10.7 GB spare on a RTX 5080. There is room to raise the context length or move up a quantisation level.

Generation226tok/s
Prompt processing5470tok/s
Max context128Ktokens
KV per 1K tokens0GB

Every quantisation of Gemma 3 4B on a RTX 5080

Highlighted row is the highest quality that still fits at 8K context.

QuantisationbpwWeights TotalVerdictMax ctxtok/s
FP16 / BF16 16.00 8.0 GB 9.1 GB Runs comfortably 128K 76.1
INT8 / W8A8 8.50 4.1 GB 5.2 GB Runs comfortably 128K 143
Q8_0 (GGUF) 8.50 4.1 GB 5.2 GB Runs comfortably 128K 143
FP8 (E4M3) 8.00 4.0 GB 5.1 GB Runs comfortably 128K 146
Q6_K 6.56 3.3 GB 4.4 GB Runs comfortably 128K 175
AWQ 4-bit 4.25 3.0 GB 4.1 GB Runs comfortably 128K 188
GPTQ 4-bit 4.25 3.0 GB 4.1 GB Runs comfortably 128K 188
MXFP4 4.25 3.0 GB 4.1 GB Runs comfortably 128K 188
Q5_K_M 5.67 2.8 GB 3.9 GB Runs comfortably 128K 200
Q5_K_S 5.52 2.8 GB 3.9 GB Runs comfortably 128K 205
Q4_K_M 4.85 2.5 GB 3.6 GB Runs comfortably 128K 226
Q4_K_S 4.58 2.4 GB 3.5 GB Runs comfortably 128K 235
Q4_0 4.55 2.4 GB 3.4 GB Runs comfortably 128K 236
IQ4_XS 4.25 2.2 GB 3.3 GB Runs comfortably 128K 248
Q3_K_M 3.91 2.1 GB 3.2 GB Runs comfortably 128K 263
IQ3_M 3.70 2.0 GB 3.1 GB Runs comfortably 128K 273
IQ3_XXS 3.06 1.7 GB 2.8 GB Runs comfortably 128K 309
Q2_K 2.63 1.5 GB 2.6 GB Runs comfortably 128K 339
IQ2_XXS 2.06 1.3 GB 2.4 GB Runs comfortably 128K 389
IQ1_M 1.75 1.2 GB 2.3 GB Runs comfortably 128K 423

Also worth checking

Gemma 3 4B on every GPU
Full hardware tableEverything that runs on a RTX 5080
Full model tableWhy the KV cache is the part that bites
Guide