@meta
  v: 1
  route: /model/llama-3-1-8b
  generated: 2026-08-01T10:53:17.190Z

@intent
  purpose:    Show what hardware can run Llama 3.1 8B.
  audience:   ai-engineer, self-hoster, model-evaluator
  capability: inspect_model_specs, find_compatible_gpus, compare_similar_models, open_huggingface

@state
  slug: llama-3-1-8b
  name: Llama 3.1 8B
  hf_repo: meta-llama/Llama-3.1-8B-Instruct
  params_b: 8
  active_params_b: ~
  family: Meta
  type: dense
  context_k: 128
  fp16_gb: 16
  summary: Meta's small workhorse. Runs comfortably on any 8GB+ card at Q4 or any 12GB+ at FP16.
  smallest_fitting_gpu_q4: rtx-3070
  smallest_fitting_gpu_fp16: rtx-3090

@actions
  - id: open_huggingface
    method: GET
    href: https://huggingface.co/meta-llama/Llama-3.1-8B-Instruct
  - id: view_compatible_gpus
    method: GET
    href: /model/llama-3-1-8b
  - id: compare_similar
    method: GET
    href: /model/granite-8b
  - id: view_calculator
    method: GET
    href: /#calculator

@context
  > Meta's small workhorse. Runs comfortably on any 8GB+ card at Q4 or any 12GB+ at FP16.

@nav
  self:      /model/llama-3-1-8b
  parents:   [/]
  peers:     [/model/granite-8b, /model/mistral-7b, /model/qwen-2-5-7b, /model/qwen-3-5-9b]
  drilldown: /the-math