LLM Speed Estimates for Local Model Hardware
LLM Speed is an open source software tool that helps people estimate how fast local large language models will run before buying hardware. Visitors can compare performance by GPU, Mac, or other system types, or by model, using estimated tokens per second and time to first token. The site relies on verified community benchmarks, and where direct results are missing it uses similar systems or specification-based projections to estimate real world speed. It is built for anyone choosing hardware for local AI workloads.
Domain: llmspeed.dev
Added: 2026-10-18
29
DRv3 · Ahrefs
0
DRv4 · Ahrefs
1.41
Open Page Rank (OPR)
#14,553,561
Global rank
8
Referring domains
#518,345
Majestic Million (MM)
275
Referring subnets
358
Referring IPs
Data: Majestic Million © Majestic, CC BY 3.0. Ranked values are selected from the original top-million dataset.
Visit website
0 likes