Qwen on NVIDIA Ampere

Can I run Qwen2.5 7B on an RTX 3080 10GB?

Yes. Qwen2.5 7B at Q4_K_M uses 5.6 GB of the 8.6 GB available on a RTX 3080 10GB, and runs at about 102 tokens per second. You can push the context to 62K.

Runs comfortably

5.6 GB of 8.6 GB · 66%
09 GB
Weights 4.4 GB
KV cache 0.4 GB
Runtime overhead 0.8 GB

Qwen2.5 7B at Q4_K_M leaves 3.0 GB spare on a RTX 3080 10GB. There is room to raise the context length or move up a quantisation level.

Generation102tok/s
Prompt processing1654tok/s
Max context62Ktokens
KV per 1K tokens0GB

Every quantisation of Qwen2.5 7B on a RTX 3080 10GB

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

QuantisationbpwWeights TotalVerdictMax ctxtok/s
FP16 / BF16 16.00 14.2 GB 15.4 GB Won't fit — 5.11
INT8 / W8A8 8.50 7.3 GB 8.5 GB Fits, but tight 9K 62.7
Q8_0 (GGUF) 8.50 7.3 GB 8.5 GB Fits, but tight 9K 62.7
FP8 (E4M3) 8.00 7.1 GB 8.3 GB Fits, but tight 13K 64.3
Q6_K 6.56 5.8 GB 7.1 GB Runs comfortably 36K 77.8
AWQ 4-bit 4.25 5.3 GB 6.5 GB Runs comfortably 46K 85.6
GPTQ 4-bit 4.25 5.3 GB 6.5 GB Runs comfortably 46K 85.6
MXFP4 4.25 5.3 GB 6.5 GB Runs comfortably 46K 85.6
Q5_K_M 5.67 5.0 GB 6.3 GB Runs comfortably 50K 89.3
Q5_K_S 5.52 4.9 GB 6.1 GB Runs comfortably 53K 91.6
Q4_K_M 4.85 4.4 GB 5.6 GB Runs comfortably 62K 102
Q4_K_S 4.58 4.2 GB 5.4 GB Runs comfortably 66K 106
Q4_0 4.55 4.2 GB 5.4 GB Runs comfortably 66K 107
IQ4_XS 4.25 3.9 GB 5.2 GB Runs comfortably 71K 113
Q3_K_M 3.91 3.7 GB 4.9 GB Runs comfortably 75K 120
IQ3_M 3.70 3.5 GB 4.8 GB Runs comfortably 78K 125
IQ3_XXS 3.06 3.0 GB 4.3 GB Runs comfortably 87K 143
Q2_K 2.63 2.7 GB 3.9 GB Runs comfortably 93K 159
IQ2_XXS 2.06 2.3 GB 3.5 GB Runs comfortably 101K 186
IQ1_M 1.75 2.0 GB 3.3 GB Runs comfortably 105K 204

Also worth checking

Qwen2.5 7B on every GPU
Full hardware tableEverything that runs on a RTX 3080 10GB
Full model tableWhy the KV cache is the part that bites
Guide