Update org card README

README.md

@@ -14,11 +14,11 @@ license: mit
     <p>
       Stable, lower-cost <strong>AI API gateway</strong> for LLM, image, and video models.
       One key for OpenAI-compatible Chat &amp; Responses, Anthropic Messages, plus native image and video contracts —
-      with ordered model-group fallbacks and per-request usage &amp; cost tracking.
+      with ordered model-group fallbacks, per-request usage &amp; cost tracking, Smart keys, and cache hit guarantee.
     </p>
     <p>
       中文：统一 AI API 网关（文本 / 图片 / 视频），OpenAI 兼容，一行切换 <code>base_url</code>，
-      带分组故障转移与按请求用量成本记录。
+      带分组故障转移、按请求用量成本、Smart keys 与缓存命中保障。
     </p>
   </div>
 
@@ -47,7 +47,7 @@ license: mit
     <div class="w-full h-28 mb-2 rounded-lg flex items-center justify-center bg-gradient-to-br from-violet-50 to-fuchsia-50 border border-gray-100 px-4">
       <span class="text-lg font-semibold text-gray-800">Start here → Console</span>
     </div>
-    <div class="underline">Hub landing · one-line SDK setup</div>
+    <div class="underline">Hub landing · bilingual</div>
   </a>
 
   <a href="https://huggingface.co/spaces/Modelflare/openai-sdk" class="block overflow-hidden group">
@@ -57,6 +57,27 @@ license: mit
     <div class="underline">One-line base_url switch</div>
   </a>
 
+  <a href="https://huggingface.co/spaces/Modelflare/coding-agents" class="block overflow-hidden group">
+    <div class="w-full h-28 mb-2 rounded-lg flex items-center justify-center bg-gradient-to-br from-indigo-50 to-pink-50 border border-gray-100 px-4">
+      <span class="text-lg font-semibold text-gray-800">Coding Agents</span>
+    </div>
+    <div class="underline">Codex · Claude Code · Gemini CLI</div>
+  </a>
+
+  <a href="https://huggingface.co/spaces/Modelflare/api-recipes" class="block overflow-hidden group">
+    <div class="w-full h-28 mb-2 rounded-lg flex items-center justify-center bg-gradient-to-br from-rose-50 to-orange-50 border border-gray-100 px-4">
+      <span class="text-lg font-semibold text-gray-800">API Recipes</span>
+    </div>
+    <div class="underline">curl · Python · endpoint picker</div>
+  </a>
+
+  <a href="https://huggingface.co/datasets/Modelflare/api-examples" class="block overflow-hidden group">
+    <div class="w-full h-28 mb-2 rounded-lg flex items-center justify-center bg-gradient-to-br from-emerald-50 to-teal-50 border border-gray-100 px-4">
+      <span class="text-lg font-semibold text-gray-800">API Examples Dataset</span>
+    </div>
+    <div class="underline">JSONL request sketches</div>
+  </a>
+
   <a href="https://docs.modelflare.dev/guides/features/?utm_source=huggingface&utm_medium=org&utm_campaign=card" class="block overflow-hidden group">
     <div class="w-full h-28 mb-2 rounded-lg flex items-center justify-center bg-gradient-to-br from-indigo-50 to-pink-50 border border-gray-100 px-4">
       <span class="text-lg font-semibold text-gray-800">Platform Features</span>
@@ -71,7 +92,8 @@ license: mit
       <li><strong>OpenAI-compatible</strong> chat — point your SDK at <code>https://modelflare.dev/v1</code></li>
       <li><strong>Group failover</strong> so one unavailable route does not fail the request</li>
       <li><strong>Per-request</strong> token, image, and video cost records</li>
-      <li>Guides for <strong>Codex, Claude Code, Gemini CLI</strong>, and other coding agents</li>
+      <li><strong>Smart keys</strong> and <strong>cache hit guarantee</strong></li>
+      <li>Guides for <strong>Codex, Claude Code, Gemini CLI</strong> — see the Coding Agents Space</li>
     </ul>
   </div>
 
@@ -86,7 +108,7 @@ client = OpenAI(
 )
 
 r = client.chat.completions.create(
-    model="grok-4.7",  # use a model id enabled on your group
+    model="grok-4.7",  # must be enabled on your group
     messages=[{"role":"user","content":"Hello from Hugging Face"}],
 )
 print(r.choices[0].message.content)</pre>
@@ -95,13 +117,17 @@ print(r.choices[0].message.content)</pre>
       Create a key in the <a href="https://modelflare.dev/?utm_source=huggingface&utm_medium=org&utm_campaign=card" class="underline">console</a>,
       then switch <code>base_url</code> to <code>https://modelflare.dev/v1</code>.
       Full guides: <a href="https://docs.modelflare.dev/?utm_source=huggingface&utm_medium=org&utm_campaign=card" class="underline">docs.modelflare.dev</a>
-      · bilingual walkthrough: <a href="https://huggingface.co/spaces/Modelflare/openai-sdk" class="underline">openai-sdk Space</a>.
+      · bilingual walkthrough: <a href="https://huggingface.co/spaces/Modelflare/openai-sdk" class="underline">openai-sdk</a>
+      · recipes: <a href="https://huggingface.co/spaces/Modelflare/api-recipes" class="underline">api-recipes</a>
+      · agents: <a href="https://huggingface.co/spaces/Modelflare/coding-agents" class="underline">coding-agents</a>.
     </p>
   </div>
 
   <div class="lg:col-span-3">
     <p>
       This Hub organization hosts Modelflare demos, examples, and community Spaces.
+      Feedback welcome in the
+      <a href="https://huggingface.co/spaces/Modelflare/start/discussions/1" class="underline">start Space discussion #1</a>.
       Interactive Gradio Chat Demo ships after HF PRO; until then use the console + API path above.
     </p>
   </div>