OpenAI on NVIDIA Blackwell

Can I run gpt-oss 20B-A3.6B on an RTX 5080?

Yes. gpt-oss 20B-A3.6B at Q4_K_M uses 13.1 GB of the 14.2 GB available on a RTX 5080, and runs at about 195 tokens per second. You can push the context to 33K.

Fits, but tight

13.1 GB of 14.2 GB · 92%
014 GB
Weights 11.9 GB
KV cache 0.4 GB
Runtime overhead 0.8 GB

This fits with almost nothing to spare. A background application claiming VRAM will push it over. Drop to the next quantisation down, or quantise the KV cache to Q8_0 — that halves the cache for no meaningful quality loss.

Generation195tok/s
Prompt processing6533tok/s
Max context33Ktokens
KV per 1K tokens0GB

Every quantisation of gpt-oss 20B-A3.6B on a RTX 5080

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

QuantisationbpwWeights TotalVerdictMax ctxtok/s
FP16 / BF16 16.00 38.9 GB 40.1 GB Won't fit — 8.76
INT8 / W8A8 8.50 20.4 GB 21.6 GB Won't fit — 26.2
FP8 (E4M3) 8.00 19.5 GB 20.6 GB Won't fit — 30.0
Q6_K 6.56 16.0 GB 17.1 GB Won't fit — 50.5
Q5_K_M 5.67 13.8 GB 15.0 GB Won't fit — 96.2
Q5_K_S 5.52 13.4 GB 14.6 GB Won't fit — 127
AWQ 4-bit 4.25 11.9 GB 13.1 GB Fits, but tight 32K 195
GPTQ 4-bit 4.25 11.9 GB 13.1 GB Fits, but tight 32K 195
MXFP4 4.25 11.9 GB 13.1 GB Fits, but tight 32K 195
Q4_K_M 4.85 11.9 GB 13.1 GB Fits, but tight 33K 195
Q4_K_S 4.58 11.3 GB 12.4 GB Runs comfortably 46K 203
Q4_0 4.55 11.2 GB 12.4 GB Runs comfortably 48K 204
IQ4_XS 4.25 10.5 GB 11.7 GB Runs comfortably 63K 214
Q3_K_M 3.91 9.7 GB 10.9 GB Runs comfortably 79K 226
IQ3_M 3.70 9.2 GB 10.4 GB Runs comfortably 89K 234
IQ3_XXS 3.06 7.8 GB 8.9 GB Runs comfortably 121K 262
Q2_K 2.63 6.8 GB 8.0 GB Runs comfortably 128K 285
IQ2_XXS 2.06 5.5 GB 6.7 GB Runs comfortably 128K 323
IQ1_M 1.75 4.8 GB 5.9 GB Runs comfortably 128K 349

Also worth checking

gpt-oss 20B-A3.6B on every GPU
Full hardware tableEverything that runs on a RTX 5080
Full model tableWhy the KV cache is the part that bites
Guide