Drop kimi-k2-thinking from the Kimi integration page

Moonshot serves newer platform accounts only kimi-k3 — K2-era ids and
the kimi-latest alias return resource_not_found_error (verified live).
Remove the K2 Thinking mentions, rewrite the dual-provider note (a
single model means one Kimi provider per account: pick the API shape
your agents use), and retake the models screenshot with the single
catalog model.

src/pages/agent-network/integrations/kimi.mdx

@@ -6,8 +6,8 @@ export const description =
 # Kimi (Moonshot AI)
 
 [Kimi](https://platform.moonshot.ai) is Moonshot AI's model family, headlined by **Kimi K3**,
-a frontier coding model with a 1M-token context window, alongside **Kimi K2 Thinking** for
+a frontier coding model with a 1M-token context window — and the only model Moonshot's
-reasoning-heavy tasks. Connecting it behind NetBird gives your agents keyless access over the
+platform currently serves to newer accounts. Connecting it behind NetBird gives your agents keyless access over the
 tunnel: NetBird holds the Moonshot API key server-side, ties every request to a real identity
 from your IdP, and applies your policies, limits, and audit on the way to Moonshot.
 
@@ -39,11 +39,11 @@ through the endpoint, so pick it based on the tools you plan to connect (see bel
 </p>
 
 <Note>
-  A provider forwards to a single upstream URL. If your team runs both OpenAI-shaped and
+  A provider forwards to a single upstream URL, and NetBird routes requests to providers
-  Anthropic-shaped agents, connect two Kimi providers — one per upstream URL — and give
+  by model name. Because Moonshot currently serves a single model (`kimi-k3`), two Kimi
-  each a distinct model list (for example `kimi-k3` on the Anthropic-shaped provider and
+  providers would claim the same model and route ambiguously — so connect one Kimi
-  `kimi-k2-thinking` on the OpenAI-shaped one). NetBird routes requests to providers by
+  provider and pick the upstream URL for the API shape your agents use. Anthropic-shaped
-  model name, so two providers claiming the same model would route ambiguously.
+  agents (Claude Code, Kimi CLI) are the most common case.
 </Note>
 
 <Note>
@@ -51,13 +51,13 @@ through the endpoint, so pick it based on the tools you plan to connect (see bel
   (and `https://api.moonshot.cn/anthropic`) instead.
 </Note>
 
-The catalog ships with **kimi-k3** (1M context) and **kimi-k2-thinking** (256K context) and
+The catalog ships with **kimi-k3** (1M context) and its default per-token prices, which
-their default per-token prices, which drive the cost figures in
+drive the cost figures in
 [Usage & Logs](/agent-network/usage-and-logs) — adjust them if your rates differ. See
 [Providers](/agent-network/providers) for details.
 
 <p>
-    <img src="/docs-static/img/agent-network/integrations/agent-network-kimi-models.png" alt="Kimi catalog models kimi-k3 and kimi-k2-thinking with per-token prices" className="imagewrapper" />
+    <img src="/docs-static/img/agent-network/integrations/agent-network-kimi-models.png" alt="Kimi catalog model kimi-k3 with per-token prices" className="imagewrapper" />
 </p>
 
 ## Create a Policy
@@ -159,8 +159,7 @@ max_context_size = 1000000
 
 ## Result
 
-Agents point at the NetBird endpoint with no key and call `kimi-k3` (or `kimi-k2-thinking`)
+Agents point at the NetBird endpoint with no key and call `kimi-k3` by name. NetBird enforces your policies, limits, and guardrails, then forwards the request
-by name. NetBird enforces your policies, limits, and guardrails, then forwards the request
 to Moonshot with the stored key. Every call is recorded in
 [Usage & Logs](/agent-network/usage-and-logs) with the caller's identity, auth group, model,
 tokens, cost, and status.