Resources connected to Hub iABOT

#1 image model on OpenRouter (Gemini 2.5 Flash Image). Fast generation with visual context. · Google

OpenAI high-fidelity image generation and editing. · OpenAI

Edit and combine images while keeping subject, lighting and style. · ByteDance Seed

Gemini 3.1 Flash Image: Pro-level quality at Flash speed. · Google

Latest OpenAI generation: better in-image text and instructions. · OpenAI

Gemini 3 Pro Image: Google’s top quality for final assets. · Google

xAI for posters, menus and ads with readable text. · xAI

Lite variant for volume and drafts at lower cost. · Google

BFL reference model for creatives, mockups and product assets. · Black Forest Labs

Ultra-fast, affordable FLUX for thumbnails and iteration. · Black Forest Labs

Top video model on OpenRouter. Speed and quality with native audio. · Google

Alibaba Wan video generation: text-to-video and image-to-video. · Alibaba

Kling Pro: longer clips with first/last-frame control. · Kuaishou

OpenAI cinematic video for high-production clips. · OpenAI

Fast Seedance for demos and low-cost drafts. · ByteDance

Kling Standard: quality/price balance with optional audio. · Kuaishou

Alibaba video from text, a start frame or references. · Alibaba

Kling O1 line for advanced video generation. · Kuaishou

Updated HappyHorse: better continuity and sharpness. · Alibaba

xAI video with references to keep character or product. · xAI

#1 by tokens on OpenRouter. Agents, wide context and multimodal. · Xiaomi

#2 on OpenRouter. Chat, extraction and logic with strong value. Also used in Hub Studio. · DeepSeek

Tencent Hunyuan/Hy model: among this week’s most used. · Tencent

Hard reasoning and complex code. Powers app building in Hub. · DeepSeek

Zhipu GLM: high OpenRouter volume for chat and agents. · Z.ai

Nemotron Ultra (free on OpenRouter): huge context at zero token cost. · NVIDIA

High-traffic MiniMax LLM for assistants and automation. · MiniMax

StepFun Flash: fast responses at scale on OpenRouter. · StepFun

Moonshot Kimi: long context and agents. · Moonshot AI

Claude Sonnet: high-quality writing and instructions. · Anthropic

Lightweight OpenAI conversational audio for high volume. · OpenAI

Microsoft Flash TTS: fast speech synthesis for products. · Microsoft

Qwen Flash TTS: natural voice at low cost. · Qwen

Affordable open-weights TTS with ES/EN voices. Hub’s default voice. · Hexgrad

Mistral speech synthesis: fast with cloning options. · Mistral

Higher-quality OpenAI conversational audio. · OpenAI

MiniMax Turbo TTS for expressive voice in apps and flows. · MiniMax

xAI text-to-speech; top of OpenRouter’s speech ranking. · xAI

Expressive Google voice with emotion tags and multi-speaker. · Google

Deepgram enterprise TTS with natural ES/EN voices. · Deepgram

OpenAI reference STT: max accuracy in tough conditions. · OpenAI

Mistral Mini transcription for speech-to-text pipelines. · Mistral

Classic Whisper via OpenRouter: compatible and reliable. · OpenAI

Fast OpenAI STT. Hub’s default transcription model. · OpenAI

Mistral audio understanding: transcription and analysis. · Mistral

OpenAI multimodal model: understands audio input (analysis / STT). · OpenAI

Mini variant for understanding audio at lower cost. · OpenAI

Multimodal Gemini: processes audio input in chat (leads audio-input rankings). · Google

Gemini 2.5 Flash with audio input to transcribe and reason. · Google

Multimodal Gemini Pro for complex audio and fine analysis. · Google

Google Lyria ~30s clips and jingles. · Google

Lyria Pro for longer structured songs. · Google

Lightweight Perplexity embeddings for indexing at scale. · Perplexity

Qwen 8B embeddings: strong semantic recall. · Qwen

Classic OpenAI for RAG and semantic search. Widely used in Hub. · OpenAI

Higher-dimension OpenAI embeddings when accuracy matters. · OpenAI

BGE-M3: multilingual, multi-granularity for RAG. · BAAI

Mistral vectors for search and retrieval. · Mistral

Qwen 4B: size/quality balance for embeddings. · Qwen

Stable Google embeddings for retrieval pipelines. · Google

Second-gen Gemini Embedding for better recall. · Google

NVIDIA Nemotron embeddings (free) for text+vision. · NVIDIA

Shows your business location on the contact page.

Enables direct payments on your site without building a backend.

Upload a PDF and extract data or answer questions about it.

Reads text from invoices or receipts uploaded through forms.

Sends simple messages or confirmations.

Sends lists, menus and media files.

Connects your assistant directly to Telegram.

Books appointments and meetings automatically.

Sends emails from your own account.

Saves documents and screenshots directly to the cloud.

Searches for and embeds videos in real time.