← Back to Models
Anthropictransformer
Claude 3.5 Haiku
20B parameters • 12.2GB VRAM (Q4 estimate) • 200,000 context
Estimate · Q4_K_M · 4k context · batch 1. For dense 7B–70B this sits ~5–8% above the Q4_K_M GGUF file. Not peak runtime VRAM and not a measured bench.
Specifications
Parameters
20B
VRAM (Q4 est.)
12.2 GB
VRAM (FP16)
40 GB
Context Window
200,000
Architecture
transformer
License
Anthropic TOS
Finding cloud alternatives...
Run Claude 3.5 Haiku on Mac Mini M4
~12.2GB VRAM needed at Q4. Mac Mini M4 has 16GB — buy hardware or rent cloud.
We may earn a commission from cloud and hardware partners at no extra cost to you.
Find the cheapest GPU that can run Claude 3.5 Haiku
Compatibility Lab — check every GPU × quantization combination