Find and compare the top AI tools for On-Device. Browse features, pricing, and user ratings of all the AI tools and apps in the market.
Newest
ZONOS3-Mini is Zyphra's compact zero-shot voice-cloning text-to-speech model, released in November 2026. It is a sparse mixture-of-experts model with 100M active and 800M total
Spark-X2.6-1.7B is a compact open-weight language model designed for practical on-device AI. It supports coding, planning, tool use, reasoning, and multilingual tasks while keep
LFM3 is Liquid AI's family of compact models for specialized and on-device intelligence. Its first release includes text, vision, audio, omni, and decision variants built around
Spark-X2.6-4B is a compact language model for on-device agents, code generation, and general-purpose reasoning. It preserves the preceding 4B architecture while improving post-t
LFM2.5-24B-A2B is Liquid AI's reasoning model for fast, private agents running on local hardware. A hybrid architecture combines short convolutions, grouped-query attention, and
EmbeddingGemma 2 is Google's lightweight model for representing multiple content modalities in one embedding space. It extends text embeddings to combinations of images, audio,
Whistle is Cactus's compact speech-recognition model for phones, wearables, and other CPU-based devices. A 16.9 MB quantized model runs in the same C++ engine as Needle and can
MiniCPM5-2B is a compact language model for local assistants, coding, reasoning, and tool-using agents. It targets on-device and resource-constrained deployments while retaining
Muse Glimmer is Meta's open 30-billion-parameter model for always-on local agents. It is optimized for tool use, local coding, multimodal understanding, long-context memory, and
Needle 2 is a compact agentic language model for tool calling, device use, and structured extraction on tiny devices. The complete model is a 14 MB binary that runs in about 28
Bonsai 27B is PrismML's low-bit multimodal language model family based on Qwen3.6 27B and designed to run locally on consumer hardware, including phones. The release includes te
MobileWan is a Qualcomm AI Research project that brings a 5B-scale video diffusion model onto memory-constrained mobile hardware. Starting from Wan2.2-5B, it targets high-qualit
Voibe is a private AI dictation and meeting notes app for Mac and Windows that turns voice into text in any app and captures meeting notes without sending a bot into the call, b
MiniCPM-V 4.6 is a compact multimodal language model for efficient image and video understanding on edge and mobile devices. It is built for scenarios where users need strong vi
Mumble AI is a voice-first workspace for Mac and iOS that turns voice into notes, summaries, and text β recording meetings without a bot joining the call, capturing voice notes,
The Speechmatics API empowers developers to build sophisticated voice AI agents with unparalleled speed and accuracy. Offering sub-second, speaker-aware speech-to-text capabilit
Stable Audio Open Small is a 341 million parameter text-to-audio model optimized to run entirely on Arm CPUs. It is designed for quickly generating short audio samples and can p
EdgeTAM is a research-driven framework developed by Facebook Research that focuses on efficient and accurate semantic segmentation, particularly tailored for edge devices. The c
AutoBrowser is an AI agent that actually uses your browser, running in Chrome's side panel. Users describe a task in plain English and the agent reads the page, then clicks, typ
Explainium (explainium.com) is a Chrome extension that makes complex text understandable in place: select any word, sentence or paragraph on a webpage or PDF and it gives an ins