Celesto Claims CI Runner With 96% Lower Per-Minute Cost Than GitHub Actions
aniketmaurya · x · 2026-10-19
Developer aniketmaurya pitches Celesto, a new CI runner aimed at the AI-agent era where "coding agents open pull requests faster than teams can review them," claiming 96% lower per-minute cost than GitHub's standard runner.
Benchmark charts (4 vCPU / 16 GiB RAM, same capacity) show Celesto Large ahead on CPU: 233.2 vs 199.4M iterations/s single-core and 927.5 vs 786.7 four-core (17-18% faster), but slower on cache ops — 6.23s vs 3.30s for a 512 MiB cache restore and 5.18s vs 2.50s for save. Faster and cheaper for CPU-bound jobs, with trade-offs for cache-heavy workflows.
More from coding & agent
- Anthropic's Boris Cherny: 90% of Its Engineers Already Use Self-Improving AI Agents — nikola_mr64990 · 2026-10-19
- Google Research's harness² Applies RSI Techniques to the Agent Harness with a Fixed Model — LangChain · 2026-10-19
- Claude reportedly has a hidden 'Sites' feature that builds and publishes websites via chat — nikola_mr64990 · 2026-10-19
- Grok Bot hands-on: credit-card shopping agent with permissions, approval gates and purchase guarantees — XFreeze · 2026-10-19
- Agent can't find installed library, hand-builds a Gibbs sampler in NumPy instead — davidmanheim · 2026-10-19
- A year running a multi-model router: 5 design choices that survived, 3 that died — AIGODSEND · 2026-10-19