OpenCone vs CoreAI Zoo

OpenCone turns the documents on your iPhone into answers you can check. Bring your own OpenAI and Pinecone keys: OpenCone reads and splits your files on the phone, stores them in your own Pinecone index, and answers your questions with OpenAI, citing the passage behind each claim. ASK • Answers stream in as they're written, with headings, lists, tables and code. • Claims carry tags such as [S1]. Tap one to read the exact passage, with its file, page and score. • Follow-up questions see your earlier exchanges. • Ask by voice, with Apple's speech recognition. SEARCH ONE INDEX OR ALL OF THEM • Auto reads each question and searches the indexes and namespaces likely to hold the answer, up to five at once. Ask it to compare two and it searches both. • Everything searches every namespace of every index you include. • One index searches the index you pick, in one namespace or in each of them. • Questions are embedded with the model that built each index, so the indexes you already have work as they are. • Optional reranking with Pinecone's hosted rerankers. YOUR DOCUMENTS • Import PDFs, plain text, Markdown, RTF, HTML, JSON, XML, CSV, code and images. Text in images is recognized on the phone. • Choose the index and namespace for each upload, follow its progress, and see how long each step took: read, split, embed and store. • Create and delete indexes and namespaces, with the cloud, region and metric you choose. ANSWER SETTINGS THAT FOLLOW THE MODEL • Choose from the current GPT models, or enter any model ID your key can use. • Reasoning effort, answer length up to the model's maximum, detail, service tier, web search and code interpreter, each offered where the model supports it. • Add your own instructions to every answer. SEE EVERY API CALL • Settings lists every OpenAI and Pinecone endpoint OpenCone calls: what it's for, the settings that shape it, and how this session's requests went. PRIVATE BY DESIGN • Your keys stay in the iPhone's Keychain. • No account, no analytics, no ads, and no server in between: requests go straight from your iPhone to OpenAI and Pinecone. • Answers are requested with storage off, so OpenAI keeps no conversation between questions. • One reset removes your keys, your settings and what OpenCone learned about your indexes. OpenCone works with your own OpenAI API key and your own Pinecone API key and project. OpenAI and Pinecone bill their usage to your accounts.

CoreAI Zoo puts the open-weight AI model zoo on your iPhone, iPad and Mac. Pick a model, download it once, and run it entirely on your device. No account, no server, and nothing you type, say or photograph ever leaves the device. CHAT Stream answers from open language models such as Qwen3, Gemma, LFM2.5, Granite and Mistral, from 0.6B to 30B+ parameters depending on your device's memory. Vision-language models (Qwen3-VL, MiniCPM-V, LFM2.5-VL) answer questions about your photos. VISION Ask questions about an image, read text with on-device OCR, and describe what the camera sees. AUDIO Transcribe speech with Whisper large-v3-turbo, Parakeet and Qwen3-ASR. Talk to an audio-understanding model, generate speech with Kokoro and VoxCPM, and try on-device music generation. CAMERA Live depth estimation, object detection (YOLOX, RF-DETR), action recognition and super-resolution on the camera feed. TOOLS Experimental models such as diffusion-based language models. BENCH Measure load time, tokens per second and memory for any chat model on your own hardware. HOW IT WORKS Every model runs through Core ML on Apple silicon. Weights are downloaded from the Hugging Face Hub on first use over HTTPS, then cached, so the app works offline afterwards. One model is resident at a time to stay within memory limits. New models arrive through a live catalog without an app update. GETTING STARTED Start with Qwen3-0.6B (454 MB) for the fastest first run. Larger models state the free space and memory they need; a recent iPhone Pro or an Apple silicon Mac runs the mid-size ones comfortably. Model weights follow their own licenses, listed on each model's Hugging Face page. The app itself is BSD-3-Clause and part of the open CoreAI model zoo. Report issues and request models at github.com/john-rocky/coreai-model-zoo.