atomic.chat

@atomic_chat_hq

Local AI chat and Inference Engine. Enhanced by TurboQuant. Team: @gladkos @skinbagwbones @AlexFromAtomic @danyurkin @worthant_

California, USA

atomic.chat

Joined March 2026

Pinned
atomic.chat
@atomic_chat_hq
22h
New Muse Glimmer 30B destroyed Gemma 4 31B at making retro arcade games! We gave three models the same task and compared one-shot outputs. Muse Glimmer 30B, Gemma 4 31B and Qwen3.6 27B each ran locally on its own RTX 5090 with 32GB VRAM Tasks: - Space Invaders - Tetris -
00:00
AI at Meta
@AIatMeta
Aug 10
Introducing Muse Glimmer, an open-weight 30B-parameter model optimized for local, always-on agent workflows. Muse Glimmer delivers strong performance on key agentic use cases and benchmarks compared with leading models in its size category, and is designed to run entirely on
atomic.chat
@atomic_chat_hq
4h
Run Meta's new Muse Glimmer 30B♾locally with 16GB VRAM! We ship our own GGUF quants. AD-IQ3_XXS does 62 tokens/s on a single RTX 4080 with vision and DFlash, and picks the same next token as the BF16 original 90% of the time! Run the model via Atomic Chat
Mark Zuckerberg
@finkd
Aug 10
Today we're also opening the weights for Muse Glimmer, a great 30B parameter dense model that can run locally. Soon we'll also release the weights for Muse Spark 1.2, our latest foundation model. Meta is a strong supporter of open source and I'm proud of these releases. Congrats
atomic.chat
@atomic_chat_hq
Aug 6
New LFM2.5-2.6B hits DeepSeek-V4 level on tool calling and runs 3.7x faster! We ran @liquidai 's new LFM2.5-2.6B against DeepSeek-V4-Flash on one box with 4x RTX 5090. Both got the same three jobs, and each one only completes if the model fires every tool call Topics: -weather
00:00
Liquid AI
@liquidai
Aug 4
Today we release LFM2.5-2.6B, an agentic model that runs entirely on-device. It plans, calls tools, and works through multi-step tasks on phones, laptops, PCs, and robots. Data never leaves the device, and the marginal cost of each run is essentially zero. > Pre-trained on ~34T
atomic.chat
@atomic_chat_hq
Aug 5
Run Ling 3.0 Flash locally 🌀 We released GGUF quants on Hugging Face, from lossless BF16 to 1-bit, plus NVFP4! AD-Q5_K_M is the best fit for 128GB hardware (tested on DGX Spark). It matches the original's token choice 97.5% of the time and drifts 31% less than the llama.cpp
Ant Ling
@AntLingAGI
Aug 5
🚀 Today, we’re releasing INT4 and FP4 (MXFP4) variants of Ling-3.0-flash. Both run end to end on a single NVIDIA DGX Spark via our Spark-adapted SGLang path. For FP4, W4A16 is the stable default, while W4A8 is tuned for higher throughput. The efficiency and accuracy of the
atomic.chat
@atomic_chat_hq
Aug 3
Run DeepSeek V4 Flash 0731 locally 🐳 We released 14 quants on Hugging Face, from lossless BF16 to 1-bit AD-IQ2_M is the best fit for 128GB hardware. It matches the original's token choice 83.6% of the time, measured against all other V4 Flash GGUFs in the community
DeepSeek
@deepseek_ai
Jul 31
🚀 DeepSeek-V4-Flash Official API is now LIVE in public beta! 🔷 We’ve massively upgraded its Agent capabilities—benchmark scores are now far surpassing the V4-Pro-Preview. Check out the massive performance leap below! 👇 🔷 The official V4-Flash now natively supports the