@meta
  v: 1
  route: /gpu/rtx-4080-s
  generated: 2026-06-10T07:25:28.154Z

@intent
  purpose:    Show what AI models fit on the RTX 4080 Super.
  audience:   ai-engineer, self-hoster, homelab-buyer
  capability: inspect_specs, find_fitting_models, compare_to_other_gpus

@state
  slug: rtx-4080-s
  name: RTX 4080 Super
  vram_gb: 16
  bandwidth_gbs: 736
  fp16_tflops: 410
  budget_8k_gb: 12
  category: rtx-40
  summary: "Modest bump over 4080. Same fits, faster throughput."
  top_fitting_models[5]: qwen-3-6-27b, gemma-4-26b, mistral-small-3, gpt-oss-20b, starcoder2-15b

@actions
  - id: compare
    method: GET
    href: /gpu/rtx-4060-ti
  - id: compare
    method: GET
    href: /gpu/rtx-4070-ti-s
  - id: compare
    method: GET
    href: /gpu/rtx-4080
  - id: compare
    method: GET
    href: /gpu/rtx-5070-ti
  - id: view_calculator
    method: GET
    href: /#calculator
  - id: view_math
    method: GET
    href: /the-math

@context
  > Modest bump over 4080. Same fits, faster throughput.

@nav
  self:      /gpu/rtx-4080-s
  parents:   [/]
  peers:     [/gpu/rtx-4060-ti, /gpu/rtx-4070-ti-s, /gpu/rtx-4080, /gpu/rtx-5070-ti]
  drilldown: /the-math