RTX 5080

Model configurations

Each card is a model and hardware setup, including any documented host RAM, storage or CPU requirements. Select a recipe to update its speeds, context and exact command.

Search configurations

Sort byModelHardware setupTotal memory (GB)RecipeContextDecode (tok/s)Prefill (tok/s)Evidence↑ A–Z

5 configurations · 6 recipes

Gemma 4 26B A4B

RTX 508016 GB1 card per machine · 1 machine

TabbyAPIEXL3 3.1 bpw

99.7decode tok/s

2,127prefill tok/s

32Kcontext

Tested on this card · Sep 26, 2026

CommandFull recipe ›

Qwen3.5-9B

RTX 508016 GB1 card per machine · 1 machine

TabbyAPIEXL3 4 bpw

179.4decode tok/s

—prefill tok/s

256Kcontext

Earlier check · Sep 22, 2026

CommandFull recipe ›

Qwen3.8-27B

RTX 508016 GB1 card per machine · 1 machine

Exact recipeexllamav3 · EXL3 2.5 bpw · 172K contextexllamav3 · EXL3 2 bpw · 195K context

—decode tok/s

—prefill tok/s

172Kcontext

Reported by MiaAI-Lab · Sep 10, 2026

CommandFull recipe ›

Qwen3.8-Flash-Next

RTX 508016 GB1 card per machine · 1 machine

Host RAM

≥ 48 GB free

Disk

≥ 64 GB NVMe

SGLangEXL3 2.05 bpw

43.1decode tok/s

1,490prefill tok/s

256Kcontext

Tested on this card · Oct 16, 2026

CommandFull recipe ›

Underdog Saluki 27B 1.0

RTX 508016 GB1 card per machine · 1 machine

llama.cppFormat not recorded

73.3decode tok/s

1,553prefill tok/s

32Kcontext

Tested on this card · Oct 9, 2026

CommandFull recipe ›

Speeds are the selected recipe’s recorded measurements, not estimates; “—” means unmeasured. Evidence distinguishes checks on this card, sibling-card checks, earlier checks and publisher reports.

Sources: NVIDIA, NVIDIA (2), newegg, microcenter