what/llm/can/i/run
gpus/nvidia/rtx-4080

RTX 4080

Released 2022-11-16 · Street $1,099
vram
16 GB
discrete
bandwidth
716
GB/s
models fit
12
top · Q4_K_M
power
320 W
medium

Spec sheet · top pick

Hardware spec
RTX 4080
NVIDIA · Ada Lovelace · sm_89
vram
16 GB
bandwidth
716 GB/s
tdp
320 W
released
2022-11-16
msrp
$1,199
street
~$1,099
★ Sweet spot for this card
gemma 3 27b
Q4_K_M · GGUF · 14.7 GB
bench50.6 · conf A
vram14.7 / 16 GB
speed49 tok/s
Best benchmark per GB of VRAM in your tier. Q4_K_Mwastes <3% quality vs FP16, leaves headroom for 32k context.
Full model inspect →

Top 10 models on this GPU

#modelquantvramtok/sbench ↓fit
01gemma 3 27bQ4_K_M14.7 GB4950.6Ok
02Qwen2.5 Coder 32BQ2_K13.1 GB5548.8Great
03Gemma 2 27BQ4_K_M14.7 GB4946.2Ok
04Gemma 2 9BQ8_010 GB7235.0Great
05Qwen3 32BQ2_K13.1 GB5540.0Great
06Gemma 3 12BQ8_013.1 GB5531.3Great
07deepseek r1 distill qwen 32bQ2_K13.1 GB5533.3Great
08Gemma 3 4BFP168.9 GB8015.1Great
09Gemma 2 2bFP164.7 GB1520.0Great
10Meta Llama 3.1 8BQ8_08.9 GB800.0Great
Showing 10 of 12 models that fit · see full ranked list →

Upgrade options

★ Best value
+RTX 5090
32 GB · ~$2,399 · +$1,300 over RTX 4080
Unlocks +4 models not in current rank
See RTX 5090
Bigger tier
+RTX 4090
24 GB · ~$1,899 · +$800 over RTX 4080
Unlocks +1 models not in current rank
See RTX 4090
Premium tier
+RTX 3090 Ti
24 GB · ~$950 · +$-149 over RTX 4080
Unlocks +1 models not in current rank
See RTX 3090 Ti

Also compare

RTX 5090
32 GB · NVIDIA · 118% more
RTX 5080
16 GB · NVIDIA · 0% more
RTX 5070 Ti
16 GB · NVIDIA · 23% cheaper
RTX 5070
12 GB · NVIDIA · 45% cheaper
Not running a RTX 4080?
Tell us what you actually have — we'll re-rank everything for your machine.
Change my machine