Files
skills/public/hanzo-live.html
zeekayandHanzo Dev 9e10c1c10e Each llms index points at the other two
hanzoskills.com/llms.txt now opens with the install line, the API
reference index and the company index, so an agent that lands on any
one of the three can find the rest.

robots.txt claimed 537 documents while llms.txt counts 920. The number
is gone from robots.txt; llms.txt derives it.

public/ is regenerated from scripts/site.py, which it had fallen
behind: the copy-prompt menu and agent bar were in the generator and
not in the committed pages. tests/site.mjs passes against the result.

Co-authored-by: Hanzo Dev <[email protected]>
2026-09-11 00:27:59 -07:00

966 lines
34 KiB
HTML

<!doctype html>
<html lang="en"><head>
<meta charset="utf-8"><meta name="viewport" content="width=device-width,initial-scale=1">
<title>hanzo-live</title>
<meta name="color-scheme" content="dark">
<style>
/* Committed to one visual world: a dark instrument panel. No light variant —
the dot field and the hairline rules only read on black, and a naive
inversion would wash both out. Every colour is painted explicitly so the page
never borrows a host background. */
@font-face{font-family:'Zen';src:url('/fonts/Zen-Variable.woff2') format('woff2');font-weight:100 900;font-style:normal;font-display:swap}
@font-face{font-family:'Zen';src:url('/fonts/Zen-Italic-Variable.woff2') format('woff2');font-weight:100 900;font-style:italic;font-display:swap}
@font-face{font-family:'Zen Mono';src:url('/fonts/ZenMono-Variable.woff2') format('woff2');font-weight:100 900;font-style:normal;font-display:swap}
@font-face{font-family:'Zen Mono';src:url('/fonts/ZenMono-Italic-Variable.woff2') format('woff2');font-weight:100 900;font-style:italic;font-display:swap}
@font-face{font-family:'Zen Pixel Square';src:url('/fonts/ZenPixel-Square.woff2') format('woff2');font-style:normal;font-display:swap}
:root{
--ink:#f4f4f2; --soft:#9c9c98; --faint:#6a6a66; --ground:#000; --panel:#0a0a0a;
--line:#232323; --line-hi:#3a3a3a; --accent:#f4f4f2;
--font-zen-sans:'Zen',-apple-system,BlinkMacSystemFont,"Segoe UI",Helvetica,Arial,sans-serif;
--font-zen-mono:'Zen Mono',ui-monospace,"SF Mono",SFMono-Regular,Menlo,Consolas,monospace;
--font-zen-pixel-square:'Zen Pixel Square',monospace;
--mono:var(--font-zen-mono);
--sans:var(--font-zen-sans);
/* Controls are pills and surfaces are soft, as on hanzo.ai. */
--r:999px; --r-soft:14px;
}
*{box-sizing:border-box}
html{-webkit-text-size-adjust:100%}
body{margin:0;background:var(--ground);color:var(--ink);font-family:var(--sans);
font-size:16px;line-height:1.65;-webkit-font-smoothing:antialiased}
.wrap{max-width:60rem;margin:0 auto;padding:0 1.75rem}
/* The dot field. A halftone made of one repeating radial-gradient, faded out
along the diagonal by a mask so it reads as a wave rather than wallpaper. */
/* No overflow clip here: the dot field is inset to the box already, and a clip
would cut off the copy-prompt menu. */
.hero{position:relative;border-bottom:1px solid var(--line)}
.hero::before{content:"";position:absolute;inset:0;
background-image:radial-gradient(var(--line-hi) 1px, transparent 1px);
background-size:7px 7px;
-webkit-mask-image:linear-gradient(115deg,#000 0%,rgba(0,0,0,.55) 38%,transparent 72%);
mask-image:linear-gradient(115deg,#000 0%,rgba(0,0,0,.55) 38%,transparent 72%);
pointer-events:none}
.hero .wrap{position:relative;padding-top:5.5rem;padding-bottom:4rem}
nav{border-bottom:1px solid var(--line)}
nav .in{max-width:60rem;margin:0 auto;padding:1.1rem 1.75rem;display:flex;
justify-content:space-between;align-items:center;gap:1rem}
.brand{font-family:var(--mono);font-size:.85rem;letter-spacing:.14em;
text-transform:uppercase;color:var(--ink);text-decoration:none;
display:inline-flex;align-items:center;gap:.6rem}
.brand i{font-style:normal;color:var(--faint)}
.mark{width:1.15em;height:1.15em;display:block;flex:0 0 auto}
nav a.up{font-family:var(--mono);font-size:.72rem;letter-spacing:.14em;
text-transform:uppercase;color:var(--soft);text-decoration:none}
nav a.up:hover{color:var(--ink)}
h1{font-family:var(--mono);font-size:clamp(2rem,5.4vw,3.4rem);line-height:1.12;
letter-spacing:-.02em;margin:0 0 1.1rem;font-weight:500;text-wrap:balance}
h1 em{font-style:normal;color:var(--faint)}
.lede{color:var(--soft);font-size:1.05rem;max-width:52ch;margin:0 0 2.25rem}
.btns{display:flex;flex-wrap:wrap;gap:.7rem;margin:0 0 .5rem;align-items:center}
.btn{display:inline-flex;align-items:center;gap:.5rem;font-family:var(--mono);
font-size:.72rem;letter-spacing:.14em;text-transform:uppercase;
padding:.8rem 1.4rem;border:1px solid var(--line-hi);border-radius:var(--r);
color:var(--ink);text-decoration:none;background:transparent;cursor:pointer;
line-height:1.2;font-weight:400}
.btn:hover{background:#141414;text-decoration:none;border-color:var(--soft)}
.btn.solid{background:var(--ink);color:#000;border-color:var(--ink)}
.btn.solid:hover{background:#d8d8d4}
.btn .chev{width:.7em;height:.7em;flex:0 0 auto;transition:transform .15s ease}
.btn[aria-expanded="true"] .chev{transform:rotate(180deg)}
/* Copy-prompt selector. Each row copies the prompt for its own target. */
.copy{position:relative;display:inline-flex}
.menu{position:absolute;top:calc(100% + .5rem);left:0;z-index:20;min-width:21rem;
background:var(--panel);border:1px solid var(--line-hi);border-radius:var(--r-soft);
padding:.4rem;display:flex;flex-direction:column;
box-shadow:0 12px 32px rgba(0,0,0,.75)}
.menu[hidden]{display:none}
.mi{display:flex;align-items:center;gap:.7rem;width:100%;text-align:left;
padding:.62rem .8rem;border:0;border-radius:calc(var(--r-soft) - .35rem);
background:transparent;color:var(--ink);font-family:var(--mono);font-size:.76rem;
letter-spacing:.04em;text-decoration:none;cursor:pointer;line-height:1.3;
white-space:nowrap}
.mi:hover,.mi:focus-visible{background:#181818;outline:none}
.mi .mark{width:1.05rem;height:1.05rem;color:var(--soft)}
.mi:hover .mark{color:var(--ink)}
.mi u{margin-left:auto;color:var(--faint);text-decoration:none;font-size:.68rem;
letter-spacing:.1em;text-transform:uppercase}
/* Works-with bar: one scrolling row, name and mark per agent. */
.strip{border-bottom:1px solid var(--line);background:var(--panel)}
.strip .in{max-width:60rem;margin:0 auto;padding:.9rem 1.75rem;display:flex;
align-items:center;gap:1.25rem;font-family:var(--mono);font-size:.72rem;
letter-spacing:.1em;text-transform:uppercase;color:var(--faint)}
.strip b{color:var(--soft);font-weight:400;flex:0 0 auto;
border-right:1px solid var(--line);padding-right:1.25rem;align-self:stretch;
display:flex;align-items:center}
.bar{display:flex;align-items:center;gap:.5rem;flex:1 1 auto;min-width:0;
overflow-x:auto;scrollbar-width:none;scroll-snap-type:x proximity;
-webkit-overflow-scrolling:touch;
-webkit-mask-image:linear-gradient(90deg,#000 calc(100% - 3rem),transparent);
mask-image:linear-gradient(90deg,#000 calc(100% - 3rem),transparent)}
.bar::-webkit-scrollbar{display:none}
.agent{display:inline-flex;align-items:center;gap:.45rem;flex:0 0 auto;
padding:.4rem .85rem;border:1px solid var(--line);border-radius:var(--r);
color:var(--soft);white-space:nowrap;scroll-snap-align:start;
text-decoration:none;background:var(--ground)}
.agent:hover{color:var(--ink);border-color:var(--line-hi);text-decoration:none}
.agent .mark{width:.95rem;height:.95rem}
h2{font-family:var(--mono);font-size:1.1rem;letter-spacing:-.01em;margin:2.75rem 0 .75rem;font-weight:500}
h3{font-size:1rem;margin:2rem 0 .5rem;font-weight:600}
h4,h5,h6{font-size:.94rem;margin:1.5rem 0 .5rem;font-weight:600}
p{margin:0 0 1rem;max-width:70ch}
ul,ol{margin:0 0 1rem;padding-left:1.35rem;max-width:70ch}
li{margin:.22rem 0}
a{color:var(--ink);text-decoration:underline;text-underline-offset:3px;
text-decoration-color:var(--line-hi)}
a:hover{text-decoration-color:var(--ink)}
a:focus-visible{outline:1px solid var(--ink);outline-offset:3px}
code{font-family:var(--mono);font-size:.85em;background:var(--panel);
border:1px solid var(--line);border-radius:5px;padding:.1em .36em}
pre{background:var(--panel);border:1px solid var(--line);padding:1rem 1.1rem;
border-radius:var(--r-soft);overflow-x:auto;margin:0 0 1.25rem}
pre code{background:none;border:0;padding:0;font-size:.82rem;line-height:1.7}
blockquote{margin:0 0 1rem;padding-left:1rem;border-left:1px solid var(--line-hi);color:var(--soft)}
hr{border:0;border-top:1px solid var(--line);margin:2.5rem 0}
table{border-collapse:collapse;width:100%;margin:0 0 1.25rem;font-size:.92rem;
display:block;overflow-x:auto}
th,td{border:1px solid var(--line);padding:.5rem .7rem;text-align:left}
.curl{display:block;background:var(--panel);border:1px solid var(--line);
border-radius:var(--r-soft);padding:.95rem 1.1rem;font-family:var(--mono);
font-size:.8rem;color:var(--soft);overflow-x:auto;margin:0 0 2.5rem;
white-space:pre;line-height:1.9}
.curl b{color:var(--ink);font-weight:400}
.curl i{font-style:normal;color:var(--faint)}
/* index */
#q{width:100%;padding:.85rem 1.2rem;font-family:var(--mono);font-size:.85rem;color:var(--ink);
background:var(--panel);border:1px solid var(--line);border-radius:var(--r);margin:0 0 .6rem}
#q::placeholder{color:var(--faint);text-transform:uppercase;letter-spacing:.1em;font-size:.72rem}
#q:focus{outline:none;border-color:var(--soft)}
#count{color:var(--faint);font-family:var(--mono);font-size:.72rem;letter-spacing:.1em;
text-transform:uppercase;margin:0 0 2rem}
.cat{margin:2.5rem 0 .5rem;font-family:var(--mono);font-size:.7rem;letter-spacing:.16em;
text-transform:uppercase;color:var(--faint);border-bottom:1px solid var(--line);
padding-bottom:.45rem}
.row{display:block;padding:.6rem .9rem;margin:0 -.9rem;border-bottom:1px solid var(--line);
border-radius:var(--r-soft);text-decoration:none}
.row:hover{background:#0a0a0a}
.nm{font-family:var(--mono);font-size:.84rem;color:var(--ink);display:block}
.ds{color:var(--soft);font-size:.9rem;display:block;margin-top:.12rem}
footer{margin-top:4rem;padding:2rem 0 4rem;border-top:1px solid var(--line);
color:var(--faint);font-size:.85rem}
footer a{color:var(--soft)}
.fmap{display:grid;gap:1.75rem;grid-template-columns:repeat(auto-fit,minmax(11rem,1fr));
margin:0 0 2rem;font-size:.85rem}
.fmap b{display:block;font-family:var(--mono);font-size:.68rem;letter-spacing:.16em;
text-transform:uppercase;color:var(--faint);font-weight:400;margin:0 0 .6rem;
border-bottom:1px solid var(--line);padding-bottom:.4rem}
.fmap a{display:inline-block;color:var(--ink);text-decoration:none;margin:.3rem .35rem 0 0}
.fmap a:hover{text-decoration:underline}
.fmap div{color:var(--faint);line-height:1.5}
.fend{border-top:1px solid var(--line);padding-top:1.25rem;font-size:.82rem}
@media (max-width:640px){
.hero .wrap{padding-top:3.25rem;padding-bottom:2.75rem}
.btns{gap:.5rem}
.btn{padding:.75rem 1.15rem;font-size:.68rem}
.copy,.copy .btn{width:100%;justify-content:center}
.menu{left:0;right:0;min-width:0}
.strip .in{flex-direction:column;align-items:flex-start;gap:.6rem;padding:.9rem 1.75rem}
.strip b{border-right:0;padding-right:0}
.bar{width:100%}
}
</style>
<link rel="alternate" type="text/markdown" href="/hanzo-live.md">
<link rel="icon" href="/favicon.svg">
<link rel="canonical" href="https://hanzoskills.com/hanzo-live">
</head>
<!-- Reading this with a machine? Every page here is also plain markdown.
This one: https://hanzoskills.com/hanzo-live.md
Index: https://hanzoskills.com/llms.txt
Corpus: https://hanzoskills.com/all.md
Append .md to any path for the markdown twin. -->
<body>
<nav><div class="in">
<a class="brand" href="/"><svg class="mark" viewBox="0 0 67 67" fill="currentColor" aria-hidden="true"><path d="M22.21 67V44.6369H0V67H22.21Z"/><path d="M66.7038 22.3184H22.2534L0.0878906 44.6367H44.4634L66.7038 22.3184Z"/><path d="M22.21 0H0V22.3184H22.21V0Z"/><path d="M66.7198 0H44.5098V22.3184H66.7198V0Z"/><path d="M66.7198 67V44.6369H44.5098V67H66.7198Z"/></svg> HANZO<i>SKILLS</i></a>
<a class="up" href="/">All skills</a>
</div></nav>
<div class="wrap">
<h1>hanzo-live</h1><p class="lede">Hanzo Live provides real-time AI streaming infrastructure for building responsive, collaborative AI applications.</p><h1>Hanzo Live - Real-Time AI Streaming</h1>
<p><strong>Category</strong>: Hanzo Ecosystem <strong>Skill Level</strong>: Intermediate to Advanced <strong>Prerequisites</strong>: WebSockets, async programming, streaming concepts</p>
<h2>Overview</h2>
<p>Hanzo Live provides real-time AI streaming infrastructure for building responsive, collaborative AI applications. Unlike traditional request-response AI APIs that require waiting for complete responses, Hanzo Live delivers token-by-token streaming with sub-50ms latency, enabling fluid conversational interfaces and real-time collaboration.</p>
<p><strong>Core Philosophy</strong>: Stream everything - tokens, thoughts, tool calls, and state updates for immediate user feedback.</p>
<h2>Key Features</h2>
<h3>⚡ Ultra-Low Latency Streaming</h3>
<ul><li><strong>Sub-50ms first token</strong>: Immediate response start</li><li><strong>Token-by-token delivery</strong>: Smooth text generation</li><li><strong>WebSocket transport</strong>: Bidirectional real-time communication</li><li><strong>Edge deployment</strong>: Minimize geographic latency</li></ul>
<h3>🔄 Real-Time Collaboration</h3>
<ul><li><strong>Multi-user sessions</strong>: Share AI conversations live</li><li><strong>Presence indicators</strong>: See who&#x27;s online and typing</li><li><strong>Collaborative editing</strong>: Multiple users, one AI stream</li><li><strong>Cursor tracking</strong>: Real-time user positions</li></ul>
<h3>🎯 Advanced Streaming Patterns</h3>
<ul><li><strong>Thought streaming</strong>: See AI reasoning process</li><li><strong>Tool call streaming</strong>: Watch tool execution in real-time</li><li><strong>Progress indicators</strong>: Loading states and completion metrics</li><li><strong>Error recovery</strong>: Graceful handling of stream interruptions</li></ul>
<h3>🛠 Developer Experience</h3>
<ul><li><strong>TypeScript SDK</strong>: Type-safe streaming clients</li><li><strong>React hooks</strong>: Drop-in streaming components</li><li><strong>State management</strong>: Automatic stream state tracking</li><li><strong>Reconnection</strong>: Automatic resume on disconnect</li></ul>
<h2>Architecture</h2>
<pre><code>┌─────────────┐ WebSocket ┌──────────────┐
│ Client │ ◄─────────────────────► │ Hanzo Live │
│ (Browser) │ │ Server │
└─────────────┘ └──────────────┘
│
│ HTTP/gRPC
▼
┌──────────────┐
│ Hanzo Node │
│ (Local) │
└──────────────┘</code></pre>
<h2>Installation</h2>
<h3>Server Setup</h3>
<pre><code class="lang-bash"># Install Hanzo Live server
npm install -g @hanzo/live-server
# Or with Docker
docker pull hanzoai/live-server
# Start server
hanzo-live serve \
--port 3001 \
--node-url http://localhost:8080 \
--redis redis://localhost:6379</code></pre>
<h3>Client Installation</h3>
<pre><code class="lang-bash"># React/Next.js projects
npm install @hanzo/live
# Vanilla JavaScript
npm install @hanzo/live-client
# Python clients
pip install hanzo-live</code></pre>
<h2>Quick Start</h2>
<h3>Basic Streaming (React)</h3>
<pre><code class="lang-tsx">import { useHanzoLive } from &#x27;@hanzo/live&#x27;
function ChatApp() {
const {
messages,
isStreaming,
sendMessage
} = useHanzoLive({
url: &#x27;ws://localhost:3001&#x27;,
model: &#x27;llama-3-8b&#x27;
})
return (
&lt;div&gt;
{messages.map((msg, i) =&gt; (
&lt;div key={i}&gt;
&lt;strong&gt;{msg.role}:&lt;/strong&gt; {msg.content}
&lt;/div&gt;
))}
{isStreaming &amp;&amp; &lt;div&gt;AI is typing...&lt;/div&gt;}
&lt;input
onKeyPress={(e) =&gt; {
if (e.key === &#x27;Enter&#x27;) {
sendMessage(e.target.value)
}
}}
/&gt;
&lt;/div&gt;
)
}</code></pre>
<h3>Streaming with TypeScript SDK</h3>
<pre><code class="lang-typescript">import { HanzoLiveClient } from &#x27;@hanzo/live-client&#x27;
const client = new HanzoLiveClient({
url: &#x27;ws://localhost:3001&#x27;,
model: &#x27;llama-3-8b&#x27;
})
// Connect
await client.connect()
// Start streaming
const stream = await client.chat({
messages: [
{ role: &#x27;user&#x27;, content: &#x27;Explain quantum computing&#x27; }
]
})
// Handle tokens as they arrive
for await (const token of stream) {
process.stdout.write(token)
}</code></pre>
<h3>Streaming with Python</h3>
<pre><code class="lang-python">from hanzo_live import HanzoLiveClient
client = HanzoLiveClient(&#x27;ws://localhost:3001&#x27;)
# Synchronous streaming
for token in client.stream_chat(
messages=[
{&#x27;role&#x27;: &#x27;user&#x27;, &#x27;content&#x27;: &#x27;Write a poem&#x27;}
],
model=&#x27;llama-3-8b&#x27;
):
print(token, end=&#x27;&#x27;, flush=True)
# Async streaming
async for token in client.stream_chat_async(...):
await display_token(token)</code></pre>
<h2>Core Features</h2>
<h3>1. Token-by-Token Streaming</h3>
<pre><code class="lang-tsx">import { useStreamingCompletion } from &#x27;@hanzo/live&#x27;
function StreamingChat() {
const {
content, // Current accumulated content
isStreaming, // Is actively streaming?
tokensPerSecond, // Real-time speed metric
stop // Cancel stream
} = useStreamingCompletion({
model: &#x27;llama-3-8b&#x27;,
onToken: (token) =&gt; {
// Called for each token
console.log(&#x27;Token:&#x27;, token)
},
onComplete: (fullText) =&gt; {
// Called when stream ends
console.log(&#x27;Complete:&#x27;, fullText)
}
})
return (
&lt;div&gt;
&lt;div className=&quot;content&quot;&gt;{content}&lt;/div&gt;
{isStreaming &amp;&amp; (
&lt;div className=&quot;metrics&quot;&gt;
{tokensPerSecond.toFixed(1)} tokens/sec
&lt;button onClick={stop}&gt;Stop&lt;/button&gt;
&lt;/div&gt;
)}
&lt;/div&gt;
)
}</code></pre>
<h3>2. Thought Process Streaming</h3>
<pre><code class="lang-tsx">import { useThoughtStream } from &#x27;@hanzo/live&#x27;
function ThoughtfulChat() {
const {
thoughts, // Array of thinking steps
answer, // Final answer
isThinking // Is currently thinking?
} = useThoughtStream({
model: &#x27;qwen-2-7b&#x27;,
showThoughts: true
})
return (
&lt;div&gt;
{thoughts.map((thought, i) =&gt; (
&lt;div key={i} className=&quot;thought&quot;&gt;
&lt;span className=&quot;step&quot;&gt;Step {i+1}:&lt;/span&gt; {thought}
&lt;/div&gt;
))}
{isThinking &amp;&amp; &lt;div className=&quot;loader&quot;&gt;Thinking...&lt;/div&gt;}
{answer &amp;&amp; (
&lt;div className=&quot;answer&quot;&gt;
&lt;strong&gt;Answer:&lt;/strong&gt; {answer}
&lt;/div&gt;
)}
&lt;/div&gt;
)
}</code></pre>
<h3>3. Tool Call Streaming</h3>
<pre><code class="lang-tsx">import { useToolStream } from &#x27;@hanzo/live&#x27;
function ToolAwareChat() {
const {
messages,
toolCalls, // In-progress tool calls
toolResults, // Completed tool results
isExecuting // Is executing tools?
} = useToolStream({
model: &#x27;gpt-4&#x27;,
tools: [
{
name: &#x27;search_web&#x27;,
description: &#x27;Search the web&#x27;,
parameters: { query: &#x27;string&#x27; }
}
]
})
return (
&lt;div&gt;
{toolCalls.map((call, i) =&gt; (
&lt;div key={i} className=&quot;tool-call&quot;&gt;
&lt;span className=&quot;tool-name&quot;&gt;{call.name}&lt;/span&gt;
&lt;span className=&quot;tool-args&quot;&gt;{JSON.stringify(call.arguments)}&lt;/span&gt;
{call.status === &#x27;running&#x27; &amp;&amp; &lt;Spinner /&gt;}
{call.status === &#x27;complete&#x27; &amp;&amp; &lt;CheckMark /&gt;}
&lt;/div&gt;
))}
{toolResults.map((result, i) =&gt; (
&lt;div key={i} className=&quot;tool-result&quot;&gt;
{result.output}
&lt;/div&gt;
))}
&lt;/div&gt;
)
}</code></pre>
<h3>4. Multi-User Collaboration</h3>
<pre><code class="lang-tsx">import { useCollaborativeSession } from &#x27;@hanzo/live&#x27;
function CollaborativeChat() {
const {
messages,
users, // Connected users
userActivity, // Who&#x27;s typing, thinking, etc
sendMessage,
inviteUser
} = useCollaborativeSession({
sessionId: &#x27;project-brainstorm&#x27;,
userId: currentUser.id
})
return (
&lt;div&gt;
&lt;div className=&quot;users&quot;&gt;
{users.map(user =&gt; (
&lt;div key={user.id} className=&quot;user&quot;&gt;
&lt;Avatar src={user.avatar} /&gt;
&lt;span&gt;{user.name}&lt;/span&gt;
{userActivity[user.id] === &#x27;typing&#x27; &amp;&amp; &lt;TypingIndicator /&gt;}
&lt;/div&gt;
))}
&lt;/div&gt;
&lt;div className=&quot;messages&quot;&gt;
{messages.map((msg, i) =&gt; (
&lt;div key={i} className={`message ${msg.userId}`}&gt;
&lt;Avatar src={getUserAvatar(msg.userId)} /&gt;
&lt;div className=&quot;content&quot;&gt;{msg.content}&lt;/div&gt;
&lt;/div&gt;
))}
&lt;/div&gt;
&lt;button onClick={() =&gt; inviteUser(email)}&gt;
Invite Collaborator
&lt;/button&gt;
&lt;/div&gt;
)
}</code></pre>
<h3>5. Progress Indicators</h3>
<pre><code class="lang-tsx">import { useStreamingProgress } from &#x27;@hanzo/live&#x27;
function ProgressiveChat() {
const {
progress, // 0-100 completion percentage
estimatedTime, // ETA in seconds
tokensGenerated, // Tokens so far
totalTokens // Expected total
} = useStreamingProgress({
model: &#x27;llama-3-70b&#x27;,
maxTokens: 1000
})
return (
&lt;div&gt;
&lt;ProgressBar value={progress} max={100} /&gt;
&lt;div className=&quot;stats&quot;&gt;
&lt;span&gt;{tokensGenerated} / {totalTokens} tokens&lt;/span&gt;
&lt;span&gt;ETA: {estimatedTime}s&lt;/span&gt;
&lt;/div&gt;
&lt;/div&gt;
)
}</code></pre>
<h2>Advanced Patterns</h2>
<h3>Streaming with Error Recovery</h3>
<pre><code class="lang-typescript">import { HanzoLiveClient } from &#x27;@hanzo/live-client&#x27;
const client = new HanzoLiveClient({
url: &#x27;ws://localhost:3001&#x27;,
// Automatic reconnection
reconnect: {
enabled: true,
maxAttempts: 5,
delayMs: 1000,
exponentialBackoff: true
},
// Resume incomplete streams
resumeOnReconnect: true
})
client.on(&#x27;error&#x27;, (error) =&gt; {
console.error(&#x27;Stream error:&#x27;, error)
// Error handled, stream continues
})
client.on(&#x27;reconnect&#x27;, (attempt) =&gt; {
console.log(`Reconnecting (attempt ${attempt})...`)
})
client.on(&#x27;reconnected&#x27;, () =&gt; {
console.log(&#x27;Reconnected, resuming stream&#x27;)
})
// Start streaming with error handling
try {
for await (const token of client.stream({...})) {
display(token)
}
} catch (error) {
if (error.code === &#x27;STREAM_INTERRUPTED&#x27;) {
// Stream was interrupted but will resume
await client.waitForReconnect()
}
}</code></pre>
<h3>Streaming with Caching</h3>
<pre><code class="lang-typescript">import { HanzoLiveClient } from &#x27;@hanzo/live-client&#x27;
const client = new HanzoLiveClient({
url: &#x27;ws://localhost:3001&#x27;,
// Enable response caching
cache: {
enabled: true,
backend: &#x27;redis&#x27;,
ttl: 3600, // 1 hour
// Semantic caching (cache similar queries)
semantic: true,
similarityThreshold: 0.95
}
})
// First request streams from model
const stream1 = await client.stream({
messages: [{ role: &#x27;user&#x27;, content: &#x27;What is 2+2?&#x27; }]
})
for await (const token of stream1) {
console.log(token)
}
// Second request streams from cache (instant)
const stream2 = await client.stream({
messages: [{ role: &#x27;user&#x27;, content: &#x27;What is 2+2?&#x27; }]
})
for await (const token of stream2) {
console.log(token) // Same output, but instant
}</code></pre>
<h3>Server-Sent Events Alternative</h3>
<pre><code class="lang-typescript">// For clients that don&#x27;t support WebSockets
import { HanzoLiveSSE } from &#x27;@hanzo/live-client&#x27;
const client = new HanzoLiveSSE({
url: &#x27;https://api.hanzo.ai/live&#x27;
})
// SSE streaming (one-way, server → client)
const eventSource = await client.stream({
messages: [{ role: &#x27;user&#x27;, content: &#x27;Hello&#x27; }]
})
eventSource.addEventListener(&#x27;token&#x27;, (event) =&gt; {
console.log(&#x27;Token:&#x27;, event.data)
})
eventSource.addEventListener(&#x27;complete&#x27;, (event) =&gt; {
console.log(&#x27;Stream complete&#x27;)
eventSource.close()
})
eventSource.addEventListener(&#x27;error&#x27;, (event) =&gt; {
console.error(&#x27;Stream error:&#x27;, event)
})</code></pre>
<h3>Multiplexed Streams</h3>
<pre><code class="lang-typescript">import { HanzoLiveClient } from &#x27;@hanzo/live-client&#x27;
const client = new HanzoLiveClient({
url: &#x27;ws://localhost:3001&#x27;,
multiplexing: true // Multiple streams over one WebSocket
})
// Start multiple streams concurrently
const [stream1, stream2, stream3] = await Promise.all([
client.stream({ messages: [{ role: &#x27;user&#x27;, content: &#x27;Query 1&#x27; }] }),
client.stream({ messages: [{ role: &#x27;user&#x27;, content: &#x27;Query 2&#x27; }] }),
client.stream({ messages: [{ role: &#x27;user&#x27;, content: &#x27;Query 3&#x27; }] })
])
// Process streams in parallel
await Promise.all([
processStream(stream1, &#x27;Stream 1&#x27;),
processStream(stream2, &#x27;Stream 2&#x27;),
processStream(stream3, &#x27;Stream 3&#x27;)
])</code></pre>
<h2>Integration with Hanzo Ecosystem</h2>
<h3>With Hanzo Node (Local Streaming)</h3>
<pre><code class="lang-typescript">// Hanzo Live automatically routes to Hanzo Node for local streaming
import { HanzoLiveClient } from &#x27;@hanzo/live-client&#x27;
const client = new HanzoLiveClient({
url: &#x27;ws://localhost:3001&#x27;,
// Prefer local inference
preferLocal: true,
localNodeUrl: &#x27;http://localhost:8080&#x27;,
// Fallback to cloud if needed
fallbackToCloud: true
})
// Stream from local Hanzo Node (privacy-first)
const stream = await client.stream({
messages: [{ role: &#x27;user&#x27;, content: &#x27;Sensitive data query&#x27; }],
forceLocal: true // Never use cloud
})</code></pre>
<h3>With Hanzo MCP (Tool Streaming)</h3>
<pre><code class="lang-typescript">import { HanzoLiveClient } from &#x27;@hanzo/live-client&#x27;
const client = new HanzoLiveClient({
url: &#x27;ws://localhost:3001&#x27;,
// MCP integration for tools
mcp: {
enabled: true,
servers: [&#x27;http://localhost:8081&#x27;]
}
})
// Stream with MCP tools
const stream = await client.stream({
messages: [{ role: &#x27;user&#x27;, content: &#x27;Search the codebase&#x27; }],
tools: &#x27;auto&#x27;, // Auto-discover from MCP
streamToolCalls: true // Stream tool execution
})
for await (const event of stream) {
if (event.type === &#x27;token&#x27;) {
console.log(&#x27;Token:&#x27;, event.data)
} else if (event.type === &#x27;tool_call&#x27;) {
console.log(&#x27;Tool:&#x27;, event.tool, event.args)
} else if (event.type === &#x27;tool_result&#x27;) {
console.log(&#x27;Result:&#x27;, event.result)
}
}</code></pre>
<h3>With @hanzo/ui Components</h3>
<pre><code class="lang-tsx">import { AIChat, StreamingText } from &#x27;@hanzo/ui&#x27;
import { HanzoLiveProvider } from &#x27;@hanzo/live&#x27;
function App() {
return (
&lt;HanzoLiveProvider
url=&quot;ws://localhost:3001&quot;
model=&quot;llama-3-8b&quot;
&gt;
{/* AIChat automatically uses Hanzo Live for streaming */}
&lt;AIChat
streaming
showThoughts
enableTools
/&gt;
{/* Or use individual streaming components */}
&lt;StreamingText
source=&quot;hanzo-live&quot;
onToken={(token) =&gt; console.log(token)}
/&gt;
&lt;/HanzoLiveProvider&gt;
)
}</code></pre>
<h3>With Hanzo Python SDK</h3>
<pre><code class="lang-python">from hanzo import Hanzo
from hanzo_live import HanzoLiveServer
# Initialize Hanzo SDK for local inference
hanzo = Hanzo(inference_mode=&#x27;local&#x27;)
# Create Hanzo Live server that uses SDK
server = HanzoLiveServer(
hanzo_client=hanzo,
host=&#x27;0.0.0.0&#x27;,
port=3001,
# Enable multiplexing for concurrent streams
multiplexing=True,
max_concurrent_streams=100
)
# Custom stream handler
@server.on_stream
async def handle_stream(messages, model, **kwargs):
# Use Hanzo SDK for inference
response = hanzo.chat.completions.create(
messages=messages,
model=model,
stream=True
)
# Stream tokens to client
for chunk in response:
token = chunk.choices[0].delta.content
if token:
yield {&#x27;type&#x27;: &#x27;token&#x27;, &#x27;data&#x27;: token}
yield {&#x27;type&#x27;: &#x27;complete&#x27;}
# Start server
await server.serve()</code></pre>
<h2>Production Deployment</h2>
<h3>Docker Deployment</h3>
<pre><code class="lang-dockerfile"># Dockerfile for Hanzo Live server
FROM node:20-alpine
WORKDIR /app
# Install Hanzo Live
RUN npm install -g @hanzo/live-server
# Copy configuration
COPY hanzo-live.config.js .
# Expose WebSocket port
EXPOSE 3001
# Start server
CMD [&quot;hanzo-live&quot;, &quot;serve&quot;, \
&quot;--port&quot;, &quot;3001&quot;, \
&quot;--node-url&quot;, &quot;http://hanzo-node:8080&quot;, \
&quot;--redis&quot;, &quot;redis://redis:6379&quot;]</code></pre>
<pre><code class="lang-yaml"># docker-compose.yml
version: &#x27;3.8&#x27;
services:
hanzo-live:
build: .
ports:
- &quot;3001:3001&quot;
depends_on:
- hanzo-node
- redis
environment:
- NODE_URL=http://hanzo-node:8080
- REDIS_URL=redis://redis:6379
- MAX_CONNECTIONS=10000
- STREAM_TIMEOUT=300
hanzo-node:
image: hanzoai/node:latest
ports:
- &quot;8080:8080&quot;
volumes:
- ./models:/models
environment:
- GPU_LAYERS=auto
redis:
image: redis:7-alpine
ports:
- &quot;6379:6379&quot;</code></pre>
<h3>Kubernetes Deployment</h3>
<pre><code class="lang-yaml"># hanzo-live-deployment.yaml
apiVersion: apps/v1
kind: Deployment
metadata:
name: hanzo-live
spec:
replicas: 3
selector:
matchLabels:
app: hanzo-live
template:
metadata:
labels:
app: hanzo-live
spec:
containers:
- name: hanzo-live
image: hanzoai/live-server:latest
ports:
- containerPort: 3001
env:
- name: NODE_URL
value: &quot;http://hanzo-node:8080&quot;
- name: REDIS_URL
value: &quot;redis://redis:6379&quot;
resources:
requests:
memory: &quot;512Mi&quot;
cpu: &quot;500m&quot;
limits:
memory: &quot;2Gi&quot;
cpu: &quot;2000m&quot;
livenessProbe:
httpGet:
path: /health
port: 3001
initialDelaySeconds: 10
periodSeconds: 30
readinessProbe:
httpGet:
path: /ready
port: 3001
initialDelaySeconds: 5
periodSeconds: 10
---
apiVersion: v1
kind: Service
metadata:
name: hanzo-live
spec:
type: LoadBalancer
ports:
- port: 3001
targetPort: 3001
protocol: TCP
selector:
app: hanzo-live</code></pre>
<h3>Edge Deployment (Cloudflare Workers)</h3>
<pre><code class="lang-typescript">// Hanzo Live on Cloudflare Workers for ultra-low latency
import { HanzoLiveEdge } from &#x27;@hanzo/live-edge&#x27;
export default {
async fetch(request: Request, env: Env) {
const upgradeHeader = request.headers.get(&#x27;Upgrade&#x27;)
if (upgradeHeader === &#x27;websocket&#x27;) {
// Handle WebSocket upgrade
const server = new HanzoLiveEdge({
nodeUrl: env.HANZO_NODE_URL,
kv: env.CACHE_KV,
durable: env.SESSIONS_DO
})
return server.handleWebSocket(request)
}
return new Response(&#x27;Hanzo Live Edge&#x27;, { status: 200 })
}
}</code></pre>
<h2>Monitoring and Observability</h2>
<h3>Real-Time Metrics</h3>
<pre><code class="lang-typescript">import { HanzoLiveClient } from &#x27;@hanzo/live-client&#x27;
const client = new HanzoLiveClient({
url: &#x27;ws://localhost:3001&#x27;,
// Enable metrics collection
metrics: {
enabled: true,
endpoint: &#x27;http://prometheus:9090&#x27;
}
})
// Metrics automatically tracked:
// - hanzo_live_connections_total
// - hanzo_live_streams_active
// - hanzo_live_tokens_per_second
// - hanzo_live_latency_ms (p50, p95, p99)
// - hanzo_live_errors_total
// - hanzo_live_cache_hits_total</code></pre>
<h3>Custom Analytics</h3>
<pre><code class="lang-typescript">client.on(&#x27;stream_start&#x27;, (event) =&gt; {
analytics.track(&#x27;Stream Started&#x27;, {
model: event.model,
userId: event.userId
})
})
client.on(&#x27;stream_complete&#x27;, (event) =&gt; {
analytics.track(&#x27;Stream Complete&#x27;, {
duration: event.duration,
tokens: event.totalTokens,
tokensPerSecond: event.tokensPerSecond
})
})
client.on(&#x27;stream_error&#x27;, (event) =&gt; {
analytics.track(&#x27;Stream Error&#x27;, {
error: event.error,
model: event.model
})
})</code></pre>
<h2>Best Practices</h2>
<h3>1. Always Show Streaming Indicators</h3>
<pre><code class="lang-tsx">// ✅ Good - clear streaming feedback
{isStreaming &amp;&amp; &lt;StreamingIndicator /&gt;}
// ❌ Avoid - no user feedback
{/* No indicator */}</code></pre>
<h3>2. Handle Disconnections Gracefully</h3>
<pre><code class="lang-typescript">// ✅ Good - automatic reconnection
const client = new HanzoLiveClient({
reconnect: { enabled: true, maxAttempts: 5 },
resumeOnReconnect: true
})
// ❌ Avoid - no reconnection logic
const client = new HanzoLiveClient({})</code></pre>
<h3>3. Use Appropriate Buffer Sizes</h3>
<pre><code class="lang-typescript">// ✅ Good - balanced buffering
const client = new HanzoLiveClient({
bufferSize: 16, // Buffer 16 tokens
flushInterval: 50 // Flush every 50ms
})
// ❌ Avoid - no buffering (choppy) or too much (laggy)</code></pre>
<h3>4. Implement Proper Error States</h3>
<pre><code class="lang-tsx">// ✅ Good - comprehensive error handling
{error &amp;&amp; &lt;ErrorMessage error={error} onRetry={retry} /&gt;}
// ❌ Avoid - silent failures
{/* No error display */}</code></pre>
<h3>5. Clean Up Resources</h3>
<pre><code class="lang-typescript">// ✅ Good - proper cleanup
useEffect(() =&gt; {
const client = new HanzoLiveClient({...})
return () =&gt; client.disconnect()
}, [])
// ❌ Avoid - memory leaks
// (no cleanup on unmount)</code></pre>
<h2>Next Steps</h2>
<ol><li><strong>Read @hanzo/ui Documentation</strong> - Streaming UI components</li><li><strong>Explore Hanzo Node</strong> - Local inference for streaming</li><li><strong>Check Hanzo MCP</strong> - Tool integration in streams</li><li><strong>See WebSocket Guide</strong> - Advanced WebSocket patterns</li></ol>
<h2>Related Skills</h2>
<ul><li><strong>hanzo-ui.md</strong> - React streaming components</li><li><strong>hanzo-node.md</strong> - Local inference infrastructure</li><li><strong>hanzo-mcp.md</strong> - Tool calling in streams</li><li><strong>realtime/INDEX.md</strong> - WebSocket and SSE patterns</li></ul>
<hr>
<p><strong>Remember</strong>: Hanzo Live enables responsive, collaborative AI experiences with ultra-low latency streaming - use it for any real-time AI application.</p>
<footer>
<div class="fmap">
<div><b>Build</b>
<a href="https://hanzo.app">hanzo.app</a> app builder
<a href="https://hanzo.codes">hanzo.codes</a> code with an agent
<a href="https://hanzoskills.com">hanzoskills.com</a> what an agent reads</div>
<div><b>Run</b>
<a href="https://console.hanzo.ai">console.hanzo.ai</a> cloud console
<a href="https://api.hanzo.ai">api.hanzo.ai</a> the API</div>
<div><b>Account</b>
<a href="https://hanzo.id">hanzo.id</a> sign in
<a href="https://pay.hanzo.ai">pay.hanzo.ai</a> top up
<a href="https://billing.hanzo.ai">billing.hanzo.ai</a> invoices and usage</div>
<div><b>Company</b>
<a href="https://hanzo.ai">hanzo.ai</a> canonical
<a href="https://hanzo.ventures">hanzo.ventures</a> ventures</div>
</div>
<div class="fend"><a href="/hanzo-live.md">This page as plain text</a> · <a href="/llms.txt">Index for agents</a></div>
</footer>
</div></body></html>