quickstart →

RunsOn m9g.xlarge · EBS gp3 vs Blacksmith 4 vCPU arm64

RunsOn does each build for 3.2–6.1× less and 1.2–2.2× faster. Both 4 vCPU arm64.

All three builds 3.8× cheaper on RunsOn: $0.0231 vs $0.0876RunsOn 1.4× faster: 11m 51s vs 16m 26s
Rust build 6.1× cheaper on RunsOn: $0.0035 vs $0.0210RunsOn 2.2× faster: 1m 49s vs 3m 59s
TypeScript build 4.5× cheaper on RunsOn: $0.0024 vs $0.0111RunsOn 2.0× faster: 58s vs 1m 55s
Docker build 3.2× cheaper on RunsOn: $0.0172 vs $0.0555RunsOn 1.2× faster: 9m 04s vs 10m 32s

Side by side

Every measurement both runners share, cost first. The dot marks the better value; gap is the larger over the smaller.

Measurement RunsOn Blacksmith Gap
Cost
Cost per Rust build $0.0035 $0.0210 6.1× RunsOn
Cost per TypeScript build $0.0024 $0.0111 4.5× RunsOn
Cost per Docker build $0.0172 $0.0555 3.2× RunsOn
Price per minute $0.0016 $0.0050 3.2× RunsOn
Real builds, wall time
Rust build (spotify-⁠player) 1m 49s 3m 59s 2.2× RunsOn
TypeScript build (opencode) 58s 1m 55s 2.0× RunsOn
Docker build (PostHog) 9m 04s 10m 32s 1.2× RunsOn
CPU & memory
PassMark single-⁠thread (points) 2,470 1,353 1.8× RunsOn
PassMark CPU Mark, all cores (points) 5,615 3,783 1.5× RunsOn
sysbench CPU, 1 thread (events/s) 3,764 929 4.1× RunsOn
7-⁠Zip compression (MIPS) 41,234 17,447 2.4× RunsOn
Memory read bandwidth 28.0 GiB/s 9.0 GiB/s 3.1× RunsOn
Measurement RunsOn Blacksmith Gap
Workspace disk (fio)
Random read, 4k QD32 (IOPS) 2,994 253,175 84.6× Blacksmith
Random write, 4k QD32 (IOPS) 2,992 35,622 11.9× Blacksmith
Random read, 4k QD1 (IOPS) 1,837 32,740 17.8× Blacksmith
QD1 read latency, p99 868 µs 37 µs 23.7× Blacksmith
Sequential read 401 MiB/s 1.7 GiB/s 4.2× Blacksmith
Sequential write 401 MiB/s 829 MiB/s 2.1× Blacksmith
Sync write, 4k + fdatasync (IOPS) 543 9,366 17.2× Blacksmith
Network, cache & Docker
Download from GitHub 172 MiB/s 100 MiB/s 1.7× RunsOn
Cache restore, 4 GB (actions/cache) 16s 21s 1.3× RunsOn
Cache save, 4 GB (actions/cache) 15s 14s 1.1× Blacksmith
docker pull 4.9s 12.9s 2.6× RunsOn
Database
pgbench read/write (TPS) 3,242 2,948 1.1× RunsOn
pgbench read-⁠only (TPS) 64,684 24,073 2.7× RunsOn
Queue
Queue under burst load 18s 12s 1.5× Blacksmith
  • ≈ even: under 5% apart, or a build inside both runners' run-to-run spread
  • Queue: identical jobs queued at the same instant (burst suite)
  • RunsOn at EC2 spot, its default (7-day average of the cheapest us-east-1 zone's spot price, fetched 16 Nov 2026); 15 of its jobs fell back to on-demand (AWS had no spot capacity for the type) and are priced at the spot rate over the same billed time
  • Cost: a managed SaaS runner against RunsOn self-hosted on your AWS account; the effort of running it was not measured

Other head-to-heads