The Most Token-Efficient Open-Source AI Agent | OpenClacky
dropdown#close">
dropdown#close"<br>class="mb-4 p-2 rounded-lg transition-colors">
dropdown#close">
GitHub
Language:
locale#switch"<br>data-locale-locale-param="zh"<br>data-locale-target="button"<br>style="color: hsl(var(--color-ink-2));">中<br>locale#switch"<br>data-locale-locale-param="en"<br>data-locale-target="button"<br>style="color: hsl(var(--color-ink-2));">EN
dropdown#close" href="/creators">
Creators
dropdown#close" href="/features">
Features
dropdown#close" href="/pricing">
Pricing
dropdown#close" href="/ai-keys">
AI Keys
dropdown#close" href="/enterprise">
Enterprise
dropdown#close" href="/docs">
Docs
Sign In<br>Sign Up
30-Day Report
30 days since launch: real network-wide cache hit rate, tokens processed, and tasks completed — all public.
See live data
announcement-bar#dismiss"<br>aria-label="Dismiss announcement"<br>style="position: absolute; right: clamp(12px, 2vw, 20px); top: 50%; transform: translateY(-50%); background: transparent; border: none; color: hsl(var(--color-ink-3)); cursor: pointer; padding: 4px; display: inline-flex; align-items: center; justify-content: center; border-radius: 4px; transition: color 0.15s ease, background 0.15s ease;"<br>onmouseover="this.style.color='hsl(var(--color-ink))'; this.style.background='rgba(255,255,255,0.06)';"<br>onmouseout="this.style.color='hsl(var(--color-ink-3))'; this.style.background='transparent';">
100% Open Source<br>MIT License
clacky-ai/openclacky
The Most Token-Efficient Open-Source AI Agent
Claude Code-class capability. Lower cost than any Agent in its class.
30 days since launch · 7-day network-wide cache hit rate 94.9% · 488,097 tasks completed
Free Download
⭐ Star on GitHub<br>957
Backed by<br>MiraclePlus · ZhenFund · Yingdong Capital · Variable Capital · Sequoia China · Hillhouse Ventures
OpenClacky<br>OWNER
My sessions<br>+ New session
Weekly meeting → task breakdown
Claude Sonnet · 2 min ago
Weekly report draft
DeepSeek V4 · 3h
Daily news digest
cron · 09:00
Product group · Feishu
channel
Common skills
/deploy<br>/browser-setup<br>/skill-creator<br>/commit
Weekly meeting → task breakdown<br>Claude Sonnet 4.6 ▾
Here's today's PMF weekly meeting transcript — please extract the decisions and break them into tasks by owner.
read meeting-notes/2026-05-02-pmf.md
write tasks/2026-W18.md
Extracted 5 decisions and 8 tasks , assigned by owner: Mia 3 (interaction draft / user-interview script / event-tracking audit), Leo 3 (performance regression / rollout checklist / release schedule), Kai 2 (customer commitments follow-up / reconciliation). Results written to tasks/2026-W18.md. Want me to sync to the team calendar?
Keep talking, or drop a file…
Real Benchmarks
Cheapest. And Best.
Engineered for the best token economics — saving you the most time and money.
Lowest Cost
Head-to-head with the strongest rivals, we cost less. At equal capability, OpenClacky's token spend stays the lowest.
OpenClacky<br>US
~0.8×
16 tools · chasing 100% cache hit · smart subagent routing
Claude Code
1.0×
World-class harness, closed-source, subscription
OpenClaw
~1.5×
Peer harness agent
Hermes Agent
~3–4×
52 built-in tools, schema bloat
Cost comparisons are computed from the same model provider, using identical or capability-equivalent models billed via API.
See Full Cost Report
Top Quality
Head-to-head with the strongest rivals, output matches and details win. Scored task-by-task from first-hand reviewer comparison.
vs Claude Code & Codex
Completion
Excellent
Clarity & Layout<br>Edge
Excellent
AI Flavor (lighter is better)<br>Edge
Excellent
Output is first-tier alongside Claude Code, with a slight edge in clarity, layout and motion.
Same output, just 1/3 the cost of Codex ($4.50 vs $12.54)
Watch the Head-to-Head
Live Stats · Real Data
Don't take our word for it. Watch the network.
Not marketing copy. These numbers come from anonymous telemetry across all OpenClacky users, refreshed hourly.
Network-wide cache hit rate
Last 30 days
94.9%
7-day hit rate
What does this mean? About 85% off your input bill. Cached tokens are billed at ~10% of full price (as low as 1% on DeepSeek & MiMo) — the higher the hit rate, the more you save.
95% excellent<br>90% good<br>60–70% typical peer agents
Tokens processed yesterday
8.76B
Total tasks completed
488,097
All figures from real anonymous telemetry
Auto-refreshed every hour
What users say
Saves tokens · Works with any model · Easy to install
Real feedback from real users, unedited.
Same project, 3 hours of coding — 8× fewer tokens than Hermes.
Head, Cross-border E-commerce Company
Tried OpenClaw, Hermes, Copow — chose OpenClacky
Under 100 MB install vs Hermes at 48 GB. The WebUI is fantastic. After using this, I realized multi-agent is unnecessary. Our entire cross-border e-commerce team is on it now.
Same user
3 weeks later, full team switched
Codex with the most expensive model couldn't solve it in 30 minutes. Clacky did it with kimi-k2.6....