atomic.chat on Twitter / X

Atomic Chat: Free Local AI Chat for Mac, Windows & iPhone | Private & Offline

Free, open-source local AI chat for Mac, Windows & iPhone. Run Gemma, Qwen, DeepSeek, Llama offline. 1,000+ models, no cloud, no subscription. Download free.

Artur Chakhvadze on Twitter / X

We are releasing our first quantized checkpoints for the Qwen3.5 series of models, co-designed jointly with our inference engine to achieve maximum possible performance on Apple hardwareStarting from 0.8B, 2B and 4B modelshttps://t.co/2R8BdhAfzv— Artur Chakhvadze (@norpadon) June 8, 2026

Qwen on Twitter / X

Qwen3-TTS is officially live. We’ve open-sourced the full family—VoiceDesign, CustomVoice, and Base—bringing high quality to the open community.- 5 models (0.6B & 1.8B)- Free-form voice design & cloning- Support for 10 languages- SOTA 12Hz tokenizer for high compression-… pic.twitter.com/BSWpaYoZWj— Qwen (@Alibaba_Qwen) January 22, 2026

Unsloth AI on Twitter / X

Inference in Unsloth Studio is now ~20% faster.You can also use older pre-downloaded GGUFs from Hugging Face etc.AMD chat support for Linux now works. Data Recipes now works on macOS, AMD, CPU setups.GitHub: https://t.co/aZWYAtakBPChangelog: https://t.co/npRrgSMYs3 pic.twitter.com/cnZvoG1TeN— Unsloth AI (@UnslothAI) March 27, 2026

WebLLM Chat

Chat with AI large language models running natively in your browser

INIYSA on Twitter / X

Apple's on-device SLM, while not as strong in simple multilingual tasks as Google's Gemma 3 4B (3.3GB, 5GB in memory), seems to outperform Qwen3 4B or Phi 4 mini reasoning. It's very very impressive, especially considering its extremely reduced size of around ~1.4GB— INIYSA (@lafaiel) June 10, 2025

clem 🤗 on Twitter / X

We just released an hf CLI extension to detect the best model/quant for a user's hardware and then spins up a local coding agent. Time to go local/private/free/fast for your agents thanks to open-source! pic.twitter.com/LcVJzGCqWx— clem 🤗 (@ClementDelangue) March 17, 2026

kwindla on Twitter / X

Local voice AI with a 235 billion parameter LLM. ✅- smart-turn v2- MLX Whisper (large-v3-turbo-q4)- Qwen3-235B-A22B-Instruct-2507-3bit-DWQ- KokoroAll models running local on an M4 mac. Max RAM usage ~110GB.Voice-to-voice latency is ~950ms. There are a couple of… pic.twitter.com/iYNQlb9JkI— kwindla (@kwindla) July 27, 2025

Qwen on Twitter / X

📢Meet Qwen3.8-Max — our most capable model to date. Next week, the open weights of Qwen3.8-Max will be released, and Qwen3.8-27B is also going open-weights to meet you all!🎉Qwen3.8-Max, a new bar for coding and cowork at 2.4T parameters:- Autonomous coding: 10+ days of… pic.twitter.com/e3YFj2hqcT— Qwen (@Alibaba_Qwen) August 3, 2026

vincent on Twitter / X

Session management is working and now it will keep the model in memory instead of constantly loading. Now to fix the hardest part, gibberish issue.ALMOST THERE ~80% DONE 🔨🔨🔨Demo: @UnslothAI Llama 3.2 1B on CPU3.6 tokens/sec to about 4.1 tokens/sec (gibberish) pic.twitter.com/1tp8dXQrLo— vincent (@t0kenl1mit) August 5, 2025

Daniel Han on Twitter / X

OpenAI's OSS model possible breakdown:1. 120B MoE 5B active + 20B text only2. Trained with Float4 maybe Blackwell chips3. SwiGLU clip (-7,7) like ReLU64. 128K context via YaRN from 4K5. Sliding window 128 + attention sinks6. Llama/Mixtral arch + biasesDetails:1. 120B MoE… https://t.co/bMFp3Z6Gs5 pic.twitter.com/1NFO4utPqr— Daniel Han (@danielhanchen) August 1, 2025

@levelsio on Twitter / X

Telegram has 1 billion usersIt's the #1 or #2 chat app in most countries around the worldComplete blindspot for most AmericansAnd NOBODY uses it for its encryption (or lack of it), we use it because it has the best UX of any chat app!Also best API to develop bots with https://t.co/5k9mmJsVb2 pic.twitter.com/Up34W9Wifs— @levelsio (@levelsio) May 28, 2024