vercel · timolins · Feb 28, 2026 · Feb 28, 2026 · Feb 28, 2026 · Feb 28, 2026
diff --git a/.changeset/chat-stream-options-adapter-fallback.md b/.changeset/chat-stream-options-adapter-fallback.md
@@ -0,0 +1,5 @@
+---
+"chat": patch
+---
+
+Pass configured fallback streaming options (`updateIntervalMs` and `fallbackPlaceholderText`) through native `adapter.stream()` calls so adapters can align their fallback behavior with `Chat` streaming config.
diff --git a/.changeset/telegram-draft-streaming-polish.md b/.changeset/telegram-draft-streaming-polish.md
@@ -0,0 +1,5 @@
+---
+"@chat-adapter/telegram": patch
+---
+
+Add Telegram `sendMessageDraft` streaming in DMs with post+edit fallback, plus HTML markdown rendering for streamed and posted markdown content.
diff --git a/README.md b/README.md
@@ -53,14 +53,14 @@ See the [Getting Started guide](https://chat-sdk.dev/docs/getting-started) for a
 | Microsoft Teams | `@chat-adapter/teams` | Yes | Read-only | Yes | No | Post+Edit | Yes |
 | Google Chat | `@chat-adapter/gchat` | Yes | Yes | Yes | No | Post+Edit | Yes |
 | Discord | `@chat-adapter/discord` | Yes | Yes | Yes | No | Post+Edit | Yes |
-| Telegram | `@chat-adapter/telegram` | Yes | Yes | Partial | No | Post+Edit | Yes |
+| Telegram | `@chat-adapter/telegram` | Yes | Yes | Partial | No | Draft API (DMs) + Post+Edit | Yes |
 | GitHub | `@chat-adapter/github` | Yes | Yes | No | No | No | No |
 | Linear | `@chat-adapter/linear` | Yes | Yes | No | No | No | No |
 
 ## Features
 
 - [**Event handlers**](https://chat-sdk.dev/docs/usage) — mentions, messages, reactions, button clicks, slash commands, modals
-- [**AI streaming**](https://chat-sdk.dev/docs/streaming) — stream LLM responses with native Slack streaming and post+edit fallback
+- [**AI streaming**](https://chat-sdk.dev/docs/streaming) — stream LLM responses with native Slack/Telegram DM streaming and post+edit fallback
 - [**Cards**](https://chat-sdk.dev/docs/cards) — JSX-based interactive cards (Block Kit, Adaptive Cards, Google Chat Cards)
 - [**Actions**](https://chat-sdk.dev/docs/actions) — handle button clicks and dropdown selections
 - [**Modals**](https://chat-sdk.dev/docs/modals) — form dialogs with text inputs, dropdowns, and validation

diff --git a/apps/docs/content/docs/adapters/index.mdx b/apps/docs/content/docs/adapters/index.mdx
@@ -18,7 +18,7 @@ Adapters handle webhook verification, message parsing, and API calls for each pl
 | Edit message | ✅ | ✅ | ✅ | ✅ | ✅ | ✅ | ✅ |
 | Delete message | ✅ | ✅ | ✅ | ✅ | ✅ | ✅ | ✅ |
 | File uploads | ✅ | ✅ | ❌ | ✅ | ⚠️ Single file | ❌ | ❌ |
-| Streaming | ✅ Native | ⚠️ Post+Edit | ⚠️ Post+Edit | ⚠️ Post+Edit | ⚠️ Post+Edit | ❌ | ❌ |
+| Streaming | ✅ Native | ⚠️ Post+Edit | ⚠️ Post+Edit | ⚠️ Post+Edit | ⚠️ Draft API (DMs) + Post+Edit | ❌ | ❌ |
 
 ### Rich content
 

diff --git a/apps/docs/content/docs/adapters/telegram.mdx b/apps/docs/content/docs/adapters/telegram.mdx
@@ -141,7 +141,7 @@ TELEGRAM_API_BASE_URL=https://api.telegram.org
 | Reactions (add/remove) | Yes |
 | Cards | Text fallback + inline keyboard buttons/link buttons |
 | Modals | No |
-| Streaming | Post+Edit fallback |
+| Streaming | `sendMessageDraft` in DMs + Post+Edit fallback |
 | DMs | Yes |
 | Ephemeral messages | No |
 | File uploads | Single file (`sendDocument`) |
@@ -159,3 +159,5 @@ TELEGRAM_API_BASE_URL=https://api.telegram.org
 - `Button` and `LinkButton` in card `Actions` render as inline keyboard buttons.
 - Telegram callback data is limited to 64 bytes. Keep button `id`/`value` payloads short.
 - Other rich card elements (images/select menus/radios) render as fallback text only.
+- Native draft streaming (`sendMessageDraft`) is currently available in private chats; groups/channels use post+edit fallback.
+- `{ markdown: ... }` and `{ ast: ... }` messages are sent with Telegram HTML parse mode for formatting support; plain strings and `{ raw: ... }` are sent as plain text.
diff --git a/apps/docs/content/docs/index.mdx b/apps/docs/content/docs/index.mdx
@@ -55,7 +55,7 @@ Each adapter factory auto-detects credentials from environment variables (`SLACK
 | Microsoft Teams | `@chat-adapter/teams` | Yes | Read-only | Yes | No | Post+Edit | Yes |
 | Google Chat | `@chat-adapter/gchat` | Yes | Yes | Yes | No | Post+Edit | Yes |
 | Discord | `@chat-adapter/discord` | Yes | Yes | Yes | No | Post+Edit | Yes |
-| Telegram | `@chat-adapter/telegram` | Yes | Yes | Partial | No | Post+Edit | Yes |
+| Telegram | `@chat-adapter/telegram` | Yes | Yes | Partial | No | Draft API (DMs) + Post+Edit | Yes |
 | GitHub | `@chat-adapter/github` | Yes | Yes | No | No | No | No |
 | Linear | `@chat-adapter/linear` | Yes | Yes | No | No | No | No |
 

diff --git a/apps/docs/content/docs/streaming.mdx b/apps/docs/content/docs/streaming.mdx
@@ -62,6 +62,7 @@ await thread.post(stream);
 | Teams | Post + Edit | Posts a message then edits it as chunks arrive |
 | Google Chat | Post + Edit | Posts a message then edits it as chunks arrive |
 | Discord | Post + Edit | Posts a message then edits it as chunks arrive |
+| Telegram | Draft API (DMs) + Post + Edit | Uses `sendMessageDraft` in private chats, falls back to post+edit in other chats |
 
 The post+edit fallback throttles edits to avoid rate limits. Configure the update interval when creating your `Chat` instance:
 

diff --git a/packages/adapter-telegram/README.md b/packages/adapter-telegram/README.md
@@ -27,7 +27,57 @@ const bot = new Chat({
 });
 ```
 
-Features include mentions, reactions, typing indicators, file uploads, and card fallback rendering with inline keyboard buttons for card actions.
+Features include mentions, reactions, typing indicators, file uploads, Markdown/AST formatting via Telegram HTML parse mode, Telegram draft streaming in DMs (`sendMessageDraft`) with post+edit fallback elsewhere, and card fallback rendering with inline keyboard buttons for card actions.
+
+## Polling mode
+
+When developing locally, you typically can't expose a public URL for Telegram to send webhooks to. Polling mode uses `getUpdates` to fetch messages directly from Telegram instead — no public endpoint needed.
+
+The `longPolling` option is entirely optional. Sensible defaults are applied when omitted.
+
+```typescript
+import { createMemoryState } from "@chat-adapter/state-memory";
+
+const telegram = createTelegramAdapter({
+  botToken: process.env.TELEGRAM_BOT_TOKEN!,
+  mode: "polling",
+  // Optional — fine-tune polling behavior:
+  // longPolling: { timeout: 30, dropPendingUpdates: false },
+});
+
+const bot = new Chat({
+  userName: "mybot",
+  adapters: { telegram },
+  state: createMemoryState(),
+});
+
+// Optional manual control
+await telegram.resetWebhook();
+await telegram.startPolling();
+await telegram.stopPolling();
+```
+
+### Auto mode
+
+With `mode: "auto"` (the default), the adapter picks the right strategy for you. In a serverless environment like Vercel it uses webhooks; everywhere else (e.g. local dev) it falls back to polling.
+
+```typescript
+const telegram = createTelegramAdapter({
+  botToken: process.env.TELEGRAM_BOT_TOKEN!,
+  mode: "auto", // default
+});
+
+const bot = new Chat({
+  userName: "mybot",
+  adapters: { telegram },
+  state: createMemoryState(),
+});
+
+// Call initialize() so polling can start in long-running local processes:
+void bot.initialize();
+
+console.log(telegram.runtimeMode); // "webhook" | "polling"
+```
 
 ## Polling mode