Run Bonsai 27B locally on a 16 GB Mac 🥷 Bonsai 2 27B is @PrismML's ternary build of Qwen3.8 27B that keeps 98.2% of FP16 quality in 7 GB, it made a voxel Japanese pagoda in one prompt with Three.js! Run AI models locally -

Make product ads with @higgsfield + Blender MCPs in Atomic Chat 🎬 Connect both in the MCP hub and ask your AI to create a rough scene in Blender, then polish it into a video with Higgsfield, all in one chat Run your workflow ->

Fly plays Crossy Road by DeepSeek V4.1-Flash 🪰🐔 We ran 668 neurons from a real fruit fly brain on a MacBook, its walking neurons decide when it hops and its escape neuron makes it jump when a car looms Run AI models locally ->

Run DeepSeek V4.1 Flash NVFP4 locally 🐳 We made an NVFP4 build of DeepSeek's v4.1 flash model with @NVIDIAAI's ModelOpt recipe, every routed expert re-encoded bit-exactly for Blackwell's FP4 tensor cores and measured on 4x B200 NVFP4 build 👇

DeepSeek V4.1 Flash made DOOM in Atomic Chat 🐳 @deepseek_ai dropped a new architecture for the flash model so we tested it by making Doom in an html file and it built a real raycasting shooter with gun mechanics and enemy waves Run AI models locally ->

Run Ling 3.0 flash VL Atomic Quants via Atomic Chat 🌀 @AntLingAGI gave ling 3.0 flash a vision upgrade, now it can view and analyze images, so we passed it a flappy bird picture to re-create the game GGUF: Run AI models locally:

Running MiniCPM5-2B locally on a 16GB MacBook 💻 It beats Qwen3.5-4B on benchmarks, so we asked it to find recent AI news, it called web search, pulled back 10 results and summarized the key developments inside Atomic Chat Run local models via

Running Ling 3.0 flash Fin AD quants via Atomic Chat 🌀 It analyzed an .xlsx expenses report and built a markdown table which contains a short per-month summary of all spendings, everything locally GGUF quants:

GLM 5.3 Flash running locally makes a @Minecraft black hole mod We ran GLM 5.3 Flash Q4 locally on a rented 4x RTX PRO 6000 box and asked it to make a lightning gun mod for the real game using the Fabric API Output: -7.6M tokens -Time: ~9 hours -Decode: ~96 tok/s The mod adds a black hole rifle, when shot it spawns a black hole that pulls in blocks, with light rings that shrink all the way into the singularity, then collapses into an explosion crater that wipes out several chunks Run GLM 5.3 Flash locally in

GLM 5.3 Flash performs at GLM 5.3 level in Blender for 17x cheaper! We gave both models a live Blender over MCP and one prompt: a 2,800 sq ft duplex penthouse, double-height living room, mezzanine, floating stair, curtain wall, terrace, furnished, real PBR materials Outputs: GLM 5.3 Flash: 811 objects, 38m 52s, $0.0526 GLM 5.3: 847 objects, 40m 43s, $0.8807 GLM 5.3's prompt following was worse and it ended up making the penthouse smaller in depth, meanwhile GLM 5.3 Flash followed the prompt without any issues, along with that the flash model can be run on a 128GB MacBook 💻

Running 1-bit Qwen 3.8 Flash Next (79GB) on a MacBook Pro M5 Max 64GB at 30 tok/s 🤯 It ran a 8-minute agent loop with 6 web searches, 3 Python runs and 5 sourced tables. Run local models via Atomic Chat! GGUF:

We officially surpassed 1M model downloads on @huggingface 🔥 Your 5 most downloaded models: 1. Qwen3.8 27B — 261K 2. Ling 3.0 flash — 158K 3. Ornith 1.5 35B A3B — 50K 4. Qwen3.5 4B DFlash — 49K 5. Gemma 4 26B A4B assistant — 48K A big thanks to all of you guys for using our quants, we will keep working hard to improve them even further!

Run Ornith 1.5 9B and 35B A3B locally via Atomic Chat 🐦‍🔥 We shipped the full 9B GGUF ladder on Hugging Face, from lossless BF16 (17.9 GB) down to 2-bit (2.8 GB) and measured all against stock quants AD-Q4_K runs on a 16GB MacBook Air with 64k context and picks the same next token as the BF16 original 91.9% of the time

Run Qwen3.8 27B locally via Atomic Chat💥 We released Atomic Dynamic GGUF quants, from 8-bit (28.9 GB) down to 1-bit (8.5 GB), and measured all other Qwen3.8 GGUFs in the community AD-IQ3_S runs on a 16GB MacBook Air and picks the same next token as the BF16 original 92.4% of the time

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 default quant of the same size.

Opus 5 crushed Fable 5 at 3D destruction physics for 2x cheaper! We gave four models the same task: build three self-contained HTML scenes with real physics Prompts: - A tornado that sucks in a whole field - A wrecking ball taking down an apartment block - An overloaded truck collapsing a truss bridge Outputs: - Opus 5: 55.9K tokens, $1.40 - Fable 5: 55.1K tokens, $2.82 - Kimi K3: 35.7K tokens, $0.55 - GPT 5.6: 20.1K tokens, $0.31 Opus got all three right unlike the other models. Houses fly up the funnel and out the top, the wall breaks where the ball hits and the rubble piles up, the bridge drops the truck into the river. Fable had almost nothing on the ground for the tornado to pick up, its building collapsed on its own before the ball even touched it and its bridge blew into sticks all at once. GPT is the cheapest here but its ball never reached the building at all and its bridge fell apart in a way nothing falls apart in real life. Kimi K3, the new Chinese frontier model, ended up in the same place as GPT

OpenHands crushed Codex by 2.3× on token efficiency! We gave both agents the same task on the same model (Qwen3.5 35B): build an 8-bit Space Invaders in 3 iterations (build, fix, polish). Output: • OpenHands: 219K tokens, • Codex: 513K tokens, @OpenHandsDev beats Codex in local running. The difference is in how they handle context. OpenHands reuses the unchanged data across every pass and only pays for new tokens, while Codex re-sends and re-counts it every iteration. OpenHands finished a bit slower, but on local runs, tokens spent are the real cost.

Fable 5 totally crushed our new contest, but it cost 6x more than Opus 4.8! We gave 4 models the same prompt: build three self-contained HTML5 canvas scenes with real physics demos Prompts: — A train derailing off a broken bridge into the water — Two cars jumping off ramps and colliding mid-air over a canyon — A monster truck crushing a row of parked cars Outputs: Fable 5: 62,158 tokens, $3.12 GPT 5.5: 37,753 tokens, $1.14 Opus 4.8: 22,280 tokens, $0.56 GLM 5.2: 36,246 tokens, $0.08 Fable 5 did all three scenes at A+. The crashes looked real, things fell and broke the right way, and nothing went through the ground or floated. GPT 5.5 was the closest to Fable. In the Bigfoot show, we think GPT was even a little better. GLM 5.2 did not win any scene, but it was the cheapest by far. Fable is the best pick for quality, but you pay more for it.

Run Cline on Local AI models with Atomic Chat! @cline is a coding agent trusted by 8M+ developers. Write, refactor, ship code securely on your own hardware with local models powered by @atomic_chat_hq — no cloud, private, free and open-source

Nemotron 3 Ultra performed GPT 5.5 level 10× cheaper We gave three same prompts to build HTML5 canvas with real physics. At first scene we have water in a spinning drum. Galton board - balls through pegs into bins. And a block collision setup with extreme mass differences. Outputs: Nemotron 3 Ultra: 11.3k tokens, $0.051 GPT 5.5: 11.0k tokens, $0.57 Nemotron stays right on GPT 5.5's heels, but at 10× cheaper. The gap in quality is far smaller than the gap in price.