quickstart →

RunsOn i7i.xlarge · local NVMe vs Blacksmith 4 vCPU x64

RunsOn does each build for 2.2–3.4× less; on time, Blacksmith is faster on the Docker build and neither is clearly faster on the Rust and TypeScript builds. Both 4 vCPU x64.

All three builds 2.5× cheaper on RunsOn: $0.0301 vs $0.0741Blacksmith 1.6× faster: 8m 27s vs 13m 25s
Rust build 2.2× cheaper on RunsOn: $0.0069 vs $0.0155within run-to-run spread: 2m 54s · 1m 47s
TypeScript build 3.4× cheaper on RunsOn: $0.0036 vs $0.0121within run-to-run spread: 1m 17s · 1m 19s
Docker build 2.4× cheaper on RunsOn: $0.0196 vs $0.0465Blacksmith 1.7× faster: 5m 22s vs 9m 15s

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 $0.0069 $0.0155 2.2× RunsOn
Cost per TypeScript build $0.0036 $0.0121 3.4× RunsOn
Cost per Docker build $0.0196 $0.0465 2.4× RunsOn
Price per minute $0.0018 $0.0080 4.4× RunsOn
Real builds, wall time
Rust build (spotify-⁠player) 2m 54s 1m 47s ≈ even
TypeScript build (opencode) 1m 17s 1m 19s ≈ even
Docker build (PostHog) 9m 15s 5m 22s 1.7× Blacksmith
CPU & memory
PassMark single-⁠thread (points) 3,124 4,332 1.4× Blacksmith
PassMark CPU Mark, all cores (points) 7,781 14,771 1.9× Blacksmith
sysbench CPU, 1 thread (events/s) 2,984 6,493 2.2× Blacksmith
7-⁠Zip compression (MIPS) 23,853 28,802 1.2× Blacksmith
Memory read bandwidth 21.7 GiB/s 41.7 GiB/s 1.9× Blacksmith
Measurement RunsOn Blacksmith Gap
Workspace disk (fio)
Random read, 4k QD32 (IOPS) 150,053 728,035 4.9× Blacksmith
Random write, 4k QD32 (IOPS) 82,487 218,915 2.7× Blacksmith
Random read, 4k QD1 (IOPS) 11,192 71,728 6.4× Blacksmith
QD1 read latency, p99 281 µs 16 µs 17.7× Blacksmith
Sequential read 1,010 MiB/s 7.0 GiB/s 7.1× Blacksmith
Sequential write 788 MiB/s 1.7 GiB/s 2.2× Blacksmith
Sync write, 4k + fdatasync (IOPS) 12,431 17,925 1.4× Blacksmith
Network, cache & Docker
Download from GitHub 198 MiB/s 76 MiB/s 2.6× RunsOn
Cache restore, 4 GB (actions/cache) 10s 11s 1.2× RunsOn
Cache save, 4 GB (actions/cache) 9s 7s 1.3× Blacksmith
docker pull 5.9s 7.9s 1.3× RunsOn
Database
pgbench read/write (TPS) 7,158 12,408 1.7× Blacksmith
pgbench read-⁠only (TPS) 58,303 109,979 1.9× Blacksmith
Queue
Queue under burst load 19s 10s 1.9× 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); its jobs ran on spot
  • 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