FTMGram v3.5.1

Next-Gen Telegram MTProto & Bot API 10.3 Framework for Python

FTMGram is a rock-solid, high-performance, asynchronous Telegram MTProto client library and Bot API framework for Python. It supercharges both user accounts and bots with modern features like Turbo Multi-Worker Transfers (Upload & Download), In-Message Buttons, AI Response Token Streaming, Bot Chat History Scanning, and Telegram Stars 2.0.

⚡ Key Highlights in v3.5.1

• 🛡️ 100% Rock-Solid Stability: Zero file descriptor leaks on Linux / uvloop containers ([Errno 24] Too many open files completely eliminated).
• ⚡ Turbo Multi-Worker Upload Engine (fast_upload): Upload videos, documents, and large files up to 2GB/4GB at maximum speed using concurrent multi-worker chunk streaming.
• ⚡ Turbo Multi-Worker Download Engine (fast_download): Download media files concurrently with parallel chunk workers for gigabit saturation.
• 🤖 Bot Chat History (get_bot_chat_history): Allows bots to retrieve message history across private chats and groups using safe, high-speed 100-ID batch scanning.
• 🎨 Fluent RichMessageBuilder DSL: Construct structured rich messages with in-message buttons (<tg-button-row>), expandable quotes, tables, and paragraphs effortlessly in Python.
• 🧠 Real-Time AI Streaming (stream_text & thinking): Seamlessly stream LLM tokens (OpenAI, Anthropic, Gemini, Groq, Ollama) directly to chat drafts with animated thinking placeholders and stop buttons.
• 🧹 Batch Chat Purge (purge_messages): Safely delete hundreds of messages in seconds with automatic RPC chunking and rate-limit handling.
• 💾 Memory Media Streaming (download_media_to_memory): Download media straight into an in-memory io.BytesIO buffer without disk I/O bottlenecks.
• 👥 Multi-Client Orchestrator (MultiClient): Manage and run dozens of bots and user accounts concurrently in a single event loop.

📦 Installation

pip install -U ftmgram

Or install with fast cryptographic acceleration:

pip install -U "ftmgram[fast]"

🚀 Quick Examples

1. Fast Upload & Fast Download (Turbo Transfers)

from ftmgram import Client

app = Client("my_bot", bot_token="TOKEN")

async def upload_and_download():
    async with app:
        # Fast Upload with 8 parallel workers
        uploaded_file = await app.fast_upload("large_video.mp4", workers=8)
        msg = await app.send_video(123456789, video=uploaded_file, caption="Sent via Turbo Upload! 🚀")

        # Fast Download with 8 parallel workers
        await app.fast_download(msg, file_name="downloads/", workers=8)

app.run(upload_and_download())

2. In-Message Buttons (Bot API 10.3)

from ftmgram import Client
from ftmgram.helpers import RichMessageBuilder, Button

app = Client("my_bot", bot_token="TOKEN")

async def send_menu(chat_id: int):
    async with app:
        rich = (
            RichMessageBuilder()
            .title("💎 VIP Subscription")
            .paragraph("Select your desired membership tier:")
            .button_row(
                Button("Starter ($5)", data="plan_starter", style="primary"),
                Button("Pro ($15)", data="plan_pro", style="success"),
                Button("VIP ($25)", data="plan_vip", style="danger")
            )
            .build()
        )
        await app.send_rich_message(chat_id, rich)

app.run(send_menu(123456789))

3. Bot Chat History Retrieval

from ftmgram import Client

app = Client("my_bot", bot_token="TOKEN")

async def scan_history(chat_id: int):
    async with app:
        async for msg in app.get_bot_chat_history(chat_id, start_message_id=1, limit=50):
            print(f"[{msg.id}] {msg.text}")

app.run(scan_history(123456789))

📚 Documentation

Visit the official documentation portal: https://ftmgram.ftmbotzx.dev/

• Getting Started
• Methods Index
• Releases & Changelog

📜 License

FTMGram is licensed under the GNU Lesser General Public License v3.0 (LGPL-3.0). Copyright (C) 2024-present FTM DEVELOPERZ.