Alternative: dark benchmark diagrams (cold-start, memory, execution density)#1496
Open
NicholasKissel wants to merge 2 commits into
Open
Alternative: dark benchmark diagrams (cold-start, memory, execution density)#1496NicholasKissel wants to merge 2 commits into
NicholasKissel wants to merge 2 commits into
Conversation
An alternative take to #1494. The inline BenchmarkSection charts are refactored into three reusable dark InkPanel diagram components, sharing primitives in benchUI.tsx: - ColdStartRace: containers-vs-Agent OS cold start with a p50/p95/p99 percentile toggle (defaults to p99); the boot replays when the percentile changes. - MemoryOverhead: per-instance memory — a full 1 GiB sandbox column next to Agent OS's sliver; the workload toggle (coding agent ~131 MB <-> execution ~22 MB) grows/shrinks the fill and re-counts the multiplier. - ExecutionDensity: cost-per-execution-second as a packing story — one server packs N concurrent executions (from bench.ts at 70% utilization) vs a sandbox's one; resolves into the price, driven by workload + hardware-tier toggles. benchUI.tsx adds shared dark-surface primitives (BenchToggle segmented control, CountUpStat, BenchInfoTooltip, parseStatNumber). bench.ts exposes execs/serverMemMb/workloadMemMb/costPerHour on CostTier and trims workload labels. The HarnessArchitecture diagram from #1494 is retained. Also bundles the same pre-existing workspace changes as #1494: Navigation, Footer, registry (RegistryPageClient, registry.ts, registry-icons.ts, registry/[slug].astro), pricing page, and favicon. Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
railway-app
Bot
temporarily deployed
to
rivet-frontend / agent-os-pr-1496
|
🚅 Deployed to the agent-os-pr-1496 environment in rivet-frontend
|
…chUI) Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Summary — alternative to #1494
An alternative design for the Agent OS homepage benchmark visuals. Where #1494 keeps the inline benchmark charts and a single light cold-start card, this PR refactors the benchmark section into three reusable dark
InkPaneldiagram components sharing primitives inbenchUI.tsx. Pick one of the two PRs.New / reworked diagram components
ColdStartRace— containers-vs-Agent OS cold start with a p50 / p95 / p99 percentile toggle (defaults to p99, the most dramatic tail). Each container boots on its own (border red→green + mini bar); Agent OS boots once as one shared process. The race replays when the percentile changes.MemoryOverhead— per-instance memory: a full 1 GiB sandbox column beside Agent OS's sliver. The workload toggle (coding agent ~131 MB ⟷ execution ~22 MB) grows/shrinks the fill and re-counts the headline multiplier (8× ⟷ 47×).ExecutionDensity— cost-per-execution-second as a packing story: one server packs N concurrent executions (frombench.tsat 70% utilization) vs a sandbox's one, resolving into the price. Driven by workload + hardware-tier toggles.HarnessArchitecture— retained from Add Agent OS homepage diagrams: harness architecture + cold-start comparison #1494 (cycling agent at the center routing to Tools/Session/Sandbox/Orchestration).Supporting changes
benchUI.tsx— shared dark-surface primitives:BenchToggle(segmented control with a framer-motion layout pill),CountUpStat,BenchInfoTooltip,parseStatNumber.bench.ts—CostTiernow exposesexecs/serverMemMb/workloadMemMb/costPerHour; workload labels trimmed.AgentOSPage.tsx— the old inlineBenchColdStartChart/BenchMemoryBar/BenchCostChartare replaced by<ColdStartRace />,<MemoryOverhead />,<ExecutionDensity />(net −431 in the page).Also included
Same pre-existing workspace changes bundled in #1494:
Navigation,Footer, registry (RegistryPageClient,registry.ts,registry-icons.ts,registry/[slug].astro), pricing page, andfavicon.svg.Verification
npm run buildpasses (81 pages).🤖 Generated with Claude Code