BotConnector 2.0 • Hybrid Cloud & Local AI

One Platform for Cloud AI
and Local AI on Your Device

Your versatile AI platform for everyday productivity, web research, PDF document analysis, creative media, and coding. Use 19 free cloud models in your browser, or run open-source models locally on your machine with no cloud inference fees.

Open Free Web App → Download BCCLI (Terminal) Models & Quota

No credit card required 19 Free Models Ready Chat, Documents & Media On-Device Local AI

Local Execution on Device (Ollama) Cloud Gateway (PAYG)

● 127.0.0.1:11434 Active (Local Inference)

USER: Please analyze this draft internal report. Prepare an executive summary and key takeaways, ensuring no data leaves my machine.

Running locally on your device via Ollama (localhost). Model inference is executing on-device. Here is your executive summary:

1. Operational Summary: Core operational metrics remain robust with self-contained workflows showing immediate cost efficiency.

2. Key Finding: Processing internal documents on local models eliminated 100% of recurring token fees for routine tasks.

3. Recommendation: Keep sensitive files on local inference, and use free cloud models whenever live web research is required.

Speed: 52 tokens/sec Target: 127.0.0.1:11434 (Local) Inference Path: Localhost (On-Device)

CHOOSE YOUR WORKFLOW

Use BotConnector for Everyday Tasks and Technical Work

Open the Web App in your browser for daily tasks like chat, web research, PDF analysis, and image generation. Or open your terminal with BCCLI when you need repo-level coding.

TERMINAL AGENTFor Coding

BCCLI: Build, debug, and manage code right in your terminal

Use BCCLI when the AI needs to inspect entire repositories, understand dependencies, run shell commands, fix errors, and review git diffs without copy-pasting code.

Full project directory and repository context Run shell commands, debuggers, and test suites Automated multi-file edits with diff reviews MCP (Model Context Protocol) tool support

Download BCCLI

BOTCONNECTOR APPFor Chat & Research

Web App Chat: Discussions, live web research, and model switching

Open the web interface for technical discussions, brainstorming, and researching current information with real-time web search. Switch freely between free cloud models or models on your laptop.

Interactive chat with 19 free cloud models Real-time web search for up-to-date data Instant switching between Cloud and Local AI Secure conversation history across sessions

Open Web App Chat

BOTCONNECTOR APPFor Documents & Files

Web App Files: Deep PDF analysis, summarization, and data extraction

Upload PDFs, text documents, spreadsheet datasets, source code, or server logs. Ask AI to summarize contents, compare versions, extract key metrics, or isolate bugs without switching tabs.

Analyze PDFs, spreadsheets, documents, and logs Key takeaway extraction and executive summaries Source code review and configuration checks Direct Q&A on your uploaded document content

Analyze Documents

BOTCONNECTOR APPFor Multimedia

Web App Multimedia: Generate AI images and natural voice audio

Use the Create workspace in the app to produce high-resolution AI images and convert text into natural speech audio with realistic voice options, all from your single account.

Generate visual assets and concept illustrations Natural-sounding text-to-speech audio voices Free multimedia quota included plus PAYG options Constantly updated multimedia model catalog

Open Image & Audio Studio

Quick guidance: For daily chat, web research, PDF document analysis, and generating images or voice, use the BotConnector Web App in your browser. To edit code repositories and run test suites in your terminal, use BCCLI.

CORE CAPABILITIES

Built for Everyday Productivity and Technical Projects

From daily Q&A, real-time web research, and work document analysis to autonomous terminal coding agents.

EVERYDAY PRODUCTIVITY · ZERO SETUP

Comprehensive In-Browser Assistant for Daily Work

Simply open the Web App on your laptop or smartphone. All your everyday needs are ready in one place without complex configurations or setup steps.

Chat & Everyday Q&A Ask questions, draft emails, and work through daily tasks

Real-time Web Research Search current news and verified online data instantly

Document & PDF Analysis Upload work files or assignments for fast executive summaries

Images & Voice Synthesis Generate visuals from text and convert drafts into natural audio

Open Free Web App in Browser →

STARTER CLOUD

19 Free Cloud Models

Start with recognizable frontier and leading cloud models, including GPT-6 Luna from OpenAI, DeepSeek V4, Mistral Medium 3.5, and GLM-5.3 Flash.

GPT-6 Luna · Launch Access

DeepSeek V4

Mistral Medium 3.5

GLM-5.3 Flash

Free Starter model availability may change over time based on upstream availability and shared capacity. GPT-6 Luna is currently available as Starter Launch Access and is planned to become an included Pro benefit when subscriptions launch.

Explore Free Catalog →

ON-DEVICE · LOCAL INFERENCE

Keep Model Inference Local on Your Machine

Run model inference directly on your hardware via Ollama. Prompts and generated content on this local inference path do not need to be sent to BotConnector Cloud; cloud-only supporting features remain separate.

8 GB RAM 2.5B - 3B Models
LFM 2.5 / Mistral

          16 GB RAM
          7B - 14B Models
Qwen / DeepSeek
        

32 GB+ RAM 32B - 70B Models
Frontier Offline

Open Local AI Setup Guide →

FOR DEVELOPERS

For Developers: Terminal Coding Agent & OpenAI API

When you need technical depth, use BCCLI to inspect codebases and resolve bugs in the terminal, or point your existing tools to our single OpenAI-compatible endpoint.

$ bccli fix "unit test"

+ [test.ts] test assertions added

14 tests passed

OpenAI(base_url="https://api.botconnector.id/v1", api_key="bc_live_...")

View Developer & BCCLI Guide →

PLATFORM

BotConnector in numbers

Current BotConnector model catalog and public-access state.

61Cloud modelsBotConnector model catalog

20Free StarterAvailable to Starter accounts

42PAYG catalogAccess follows account activation

2Execution modesCloud + Local

COMPATIBILITY

Works with the tools you already use

Use the BotConnector Base URL and account API key with applications that support an OpenAI-compatible endpoint.

BCCLI BotConnector AI coding agent for the terminal OpenCode OpenAI-compatible provider Cline Custom OpenAI-compatible API Aider Custom OpenAI API base Cursor Custom Base URL for Chat/Agent Version-dependent OpenAI SDK Custom base URL OpenAI CLI Custom API base URL OpenAI-compatible Generic compatible clients

This list is limited to clients that can use the current Chat Completions contract; compatibility can depend on client version and custom Base URL support. Open integration guide →

OFFICIAL ECOSYSTEM & INFRASTRUCTURE

Direct Access to World-Class AI via Official Channels

BotConnector connects official AI vendors and large-scale inference providers through one consistent access layer for cloud, API, CLI, and local workflows.

UPSTREAM ECOSYSTEM

Official Vendors & Inference

OpenAI

Anthropic

Google

DeepSeek

Qwen

Mistral

NVIDIA

Zhipu AI

Kimi

MiniMax

BotConnector Unified AI Access Layer

Routing Direct Official Gateway

Catalog 60+ Curated Models

Protocol OpenAI Compatible

Single Access Contract

DELIVERY CHANNELS

Access Surfaces

Web App Workspace, chat, docs & media

API Drop-in OpenAI SDK integration

BCCLI Autonomous terminal coding agent

Local AI On-device hardware inference

Official API Access

Direct integrations with supported model vendors and inference providers.

Unified Access & Billing

One account, one model catalog, consistent usage and PAYG access.

Cloud + Local Freedom

Move between hosted AI and supported local models without changing the overall workflow.

ARCHITECTURE COMPARISON

Two Execution Paths: When to choose Cloud vs Local?

Complete flexibility to decide where your compute runs based on privacy and capacity requirements.

CLOUD PATH (HOSTED INFERENCE)

Massive Compute Capacity

Run large frontier models (70B+) and multimodal reasoning without purchasing dedicated GPU workstations.

19 Free Models on Starter: Instant access for daily development and testing at zero cost.
PAYG Without Forced Subscriptions: Access frontier reasoning like DeepSeek R1 billed strictly per token.
Official Model Providers: Connected directly to authorized vendor APIs with consistent quality and release parity.
Cloud Data Handling: TLS protects data in transit. Cloud requests follow the data-use and retention terms of the model or inference service used for that request.

Cloud AI Documentation →

LOCAL PATH (ON-DEVICE VIA OLLAMA)

Local Control & Zero Cloud Inference Cost

Execute models directly on your hardware. Ideal for proprietary code, internal documents, and offline workflows.

On-Device Inference: Model execution runs through localhost (127.0.0.1:11434); cloud-only supporting features remain separate.
Zero Token Fees: Unlimited local completions with no recurring monthly invoices.
Automatic Hardware Matching: Clear RAM/GPU sizing advice to prevent system thrashing.
True Offline Capability: Keep working productively during flights or connectivity outages.

Local AI Setup Guide →

CURATED MODELS

Verified Models for Every Task

A balanced catalog of local open-weights and frontier cloud models ready for production.

Qwen 2.5 Coder 14B

Coding & Refactor

Primary software engineering assistant for multi-file refactoring, unit tests, and code generation.

Local (Ollama) Recommended 16 GB RAM Q4_K_M Format

DeepSeek V3 / R1

Logic & Science

Deep mathematical reasoning, algorithmic logic verification, and complex data analysis.

Cloud (PAYG) Frontier Grade Max Capacity

LFM 2.5 2.6B / Mistral Nemo

Fast Chat

Lightweight bilingual model with high battery efficiency and rapid response times for everyday laptops.

Local / Starter Cloud Lightweight (8 GB RAM) Battery Friendly

Browse All Catalog Models →

One Platform for Cloud AIand Local AI on Your Device