LIVE TAPE
OpenClaw 88,412 stars·CVE-2026-25898 disclosed (HIGH, Hermes)·Hermes Agent v2026.4.7 published·Hermes Agent +182 stars (last hour)·OpenClaw v2026.4.6 — credential vault hardening·CVE-2026-26133 patched (NanoClaw)·Pi 5 16GB rumoured for Q3 — recheck guidance·Nanobot +47 stars (last hour)·ZeroClaw v0.4.2 — Apple container fixes·Mac Mini M4 wins quarterly hardware survey·OpenClaw 88,412 stars·CVE-2026-25898 disclosed (HIGH, Hermes)·Hermes Agent v2026.4.7 published·Hermes Agent +182 stars (last hour)·OpenClaw v2026.4.6 — credential vault hardening·CVE-2026-26133 patched (NanoClaw)·Pi 5 16GB rumoured for Q3 — recheck guidance·Nanobot +47 stars (last hour)·ZeroClaw v0.4.2 — Apple container fixes·Mac Mini M4 wins quarterly hardware survey·
PocketClawvol. 1 · 2026

What hardware do you actually need?

Pick a model size and a target tokens-per-second. We'll show which hosts hit the spec, which fall short, and the cheapest one that comfortably handles your workload.

cheapest host that meets your spec
Mac Mini M4 16 GB
749 capex · 16 W under load · 28 tok/s on a 7B model
HostCapexRAMtok/swattsverdict
Mac Mini M4 16 GB74916 GB2816✓ meets target
Single-GPU box (RTX 4060 Ti 16 GB)118032 GB9291✓ meets target
Raspberry Pi 5 8 GB1388 GB4.211below target
Intel NUC 13 i762032 GB1128below target

Methodology

Throughput numbers come from our 21-day benchmark across four hosts. See the full benchmark article for context, methodology, and the long tail of edge cases. Numbers assume Q4_K_M quantisation; FP16 will be ~2× slower with ~4× the RAM.