@meta
  v: 1
  route: /can-i-run/llama-3-2-3b
  generated: 2026-06-10T10:33:06.601Z

@intent
  purpose:    Hardware requirements to run Llama 3.2 3B locally.
  audience:   hardware-shopper, ai-engineer, self-hoster
  capability: check_compatibility, find_cheapest_fit, find_best_quality

@state
  slug: llama-3-2-3b
  name: Llama 3.2 3B
  params_b: 3.21
  fp16_gb: 6.42
  type: dense

@actions
  - id: view_model
    method: GET
    href: /model/llama-3-2-3b
  - id: view_ranked_gpus
    method: GET
    href: /best-gpu-for/llama-3-2-3b
  - id: tune_calculator
    method: GET
    href: /calc?model=llama-3-2-3b

@context
  > What hardware actually fits Llama 3.2 3B (3.21B params) across FP16 / Q8_0 / Q5_K_M / Q4_K_M / Q3_K_M quants. Includes the cheapest fit, the best-quality fit, and the Apple Silicon recommendation.

@nav
  self:      /can-i-run/llama-3-2-3b
  parents:   [/, /model/]
  peers:     [/best-gpu-for/llama-3-2-3b, /model/llama-3-2-3b]
  drilldown: /calc?model=llama-3-2-3b