Benjamin Marte (@benmarte) header
Benjamin Marte (@benmarte) profile picture

Full Stack R&D Developer Web Evangelist. Always looking to learn and improve my skills, while helping others do the same.

Tampa Fl.
Joined February 2010
Photo by Benjamin Marte (@benmarte)Photo by Benjamin Marte (@benmarte)Photo by Benjamin Marte (@benmarte)Photo by Benjamin Marte (@benmarte)Photo by Benjamin Marte (@benmarte)Photo by Benjamin Marte (@benmarte)Photo by Benjamin Marte (@benmarte)Photo by Benjamin Marte (@benmarte)Photo by Benjamin Marte (@benmarte)Photo by Benjamin Marte (@benmarte)
Benjamin Marte retweeted
This is it, everyone! We released Buzz 1.0, and what an update it is. Your people, your agents, your project, all in one place. Plugins. Make Buzz work your way. Plugins are the heart of Buzz. Add your own tools, pages or panels, extend conversations, and build the workspace you want. Share with your team or with the community. Work solo. Work together. Your own agents chats and workspaces alongside team channels. Bring others into a chat when your're ready to collaborate. Your choice of agents. Run Claude, Codex, Goos or Pi on your computer. Create teams of agents for different kinds of work. Go beyond typing. Send voice notes, jump into live audio Huddles with yuour team, or talk with supported voice-capable agents. Ask your agents to keep and eye on things. With attention, they can watch for updates, follow a PR's checks and reviews, and report back while buzz is running. On top of that we released a new version of shared compute, a plugins marketplace we will be opening up soon, agents marketplace and lots more to come. Download the new version from buzz.xyz Enjoy!
77
63
12
605
129,860
Benjamin Marte retweeted
Last Qwen3.8 Flash update with STQ 2.75bpw v2.0 pack. It moves away from hybrid affine/stq format to pure STQ with +15% quality. 64GB users can enjoy this while waiting for Qwen 4! Require Sushi 1.5.1 ro run. 👉 huggingface.co/beamster/Qwen…
7
6
51
2,752
Benjamin Marte retweeted
DeepSeek STQ 1.75bpw v2.0 is here! Custom tunings cut ~15% errors in every category, surpassing the much bigger oQ2.5e, yet runnable with 96GB+ Mac. Real demo as always, it rendered this Xmas magic weather simulations! Grab it 🤗🍣 huggingface.co/beamster/Deep… (Require 🍣 1.5.0)
12
8
1
66
4,902
Benjamin Marte retweeted
A new tiling system for macOS 1. Apps in columns 2. Clickable icon for scrolling 3. Multi-monitor support 4. Built in performance monitor 5. MCP using Apple Intelligence for layout (native) ckluis.github.io/kolom/
3
1
2
167
Benjamin Marte retweeted
You don’t need a Mac to build and install an iPad app. Today my agent built and signed an app on Linux, then installed it over Tailscale with no USB cable. No Mac in the build or install path. It's all open source: github.com/joshuaswarren/oma…
20
27
8
375
11,851
Benjamin Marte retweeted
Tutorial on how to use /print3d: - open your fav coding agent (hermes agent) - post a picture/description of what you want - wait a couple of minutes. prompt: use /print3d to make this case [pic] for my raspberry 5 but with the possibility to stack them in a row
7
1
1
72
5,749
Benjamin Marte retweeted
Replying to @nanbeige
Crazy launch 💀
3
2
20
8,702
Benjamin Marte retweeted
MLX-Serve 26.10.2 is out, download for your Mac/iOS/DGX/x86 - Same MLX Models now run everywhere. Enable Lan Sharing and share across devices. GLM 5.3, DS4.1, Mimo, MiniCPM, YuE2, and more. github.com/ddalcu/mlx-serve/…
22
16
12
189
20,741
Benjamin Marte retweeted
Sushi 🍣 1.2.2 has landed with GLM5.3 Flash support for 96GB machine! ✅ ~79GiB, 10s load time. ✅ KLD 0.103 vs BF16 (code, prose avg at 512 tokens) ✅ 512k context (8 bits) / 300k context (BF16) 🤗 huggingface.co/beamster/GLM-…
18
11
3
179
16,423
Benjamin Marte retweeted
🚨🚨🚨 keys - TensorFold Studio ONE DGX-Spark + Mac M1-M5 (64GB ram min) v2.3 Released now also with Qwen2.1 Image Turbo integrated!!!!!!!!!! Go Get It!!!!!!!!!!!!!!!!!! if you like the repo and want to support please donate to: GoFundMe in Profile github.com/drowzeys/keys-Ten…
9
9
5
96
8,527
Benjamin Marte retweeted
v1.9 First Public Release.....LFG!!!!!! repo 👇🧵
10
15
4
80
9,011
Benjamin Marte retweeted
Meet Mistral Large 4, aka Le Chonk. • 1T parameters, natively multimodal. 49B active. It is the best open weights model from US or Europe on aggregated benchmarks. • State-of-the-art on critical workloads, including cyber defense, manufacturing and finance and it surpasses closed frontier models on visual grounding. • Forged in Europe end-to-end and is deployable from Europe via our own Mistral Cloud infrastructure. • Available to all via API today. Working with cybersecurity partners privately. Open weights release end of October.
🤖 Made with AI
2,030
4,552
2,814
40,578
5,775,210
Benjamin Marte retweeted
GameToMac now supports 20 games running smoothly on Apple Silicon. Next is more launchers: Epic, GOG, Battle net and Rockstar. Then a bit more time on each game before adding new ones.
74
35
10
1,134
55,480
Benjamin Marte retweeted
Replying to @LeoBuilds_
Software engineers are supposed to do exactly what we’ve always done. Ensure that the composition and structure of the system is robust and maintainable and that the behavior of the system properly serves the needs of our users. This requires engineering discipline. It need not require detailed reading of the code; just as using compilers does not require detailed reading of the binary.
74
211
20
2,239
59,811
Benjamin Marte retweeted
If you can write CSS, you can write GSS { ● } I've been dreaming of this for years: a language to build 3D scenes like you write CSS. Selectors, cascade, :hover, keyframes… compiled to one GPU shader. Meet GSS: GPU Style Sheets. gss-lang.dev #shader #glsl #gss
92
268
58
2,187
102,085
Benjamin Marte retweeted
Six months ago I had a hunch. Local AI won't become the norm until it feels instant. Running a model on your own machine should feel like magic, not like tapping your fingers waiting for a reply. The response to TensorFold since has blown me away. Thank you to everyone testing it, sending PRs and sharing feedback. There's a lot more speed to unlock. Right now most of my time goes on reviewing and landing PRs. That matters, and every fix makes TensorFold better, but the biggest gains are in deep performance work I can't get to quickly. If you want to see what TensorFold can really do, backing would let me work on it full time, with the hardware to test faster and dig into performance, and bring it to everyone in local AI. All in the open under Apache-2.0. We have been made aware that TF is being run in real production environments, so we could potentially look into support packages. To back the project, sponsor hardware or talk about support, my DMs are open. Even just sharing this post helps. 🫶🏼
43
67
9
351
27,999
Benjamin Marte retweeted
MLX-Serve 26.10.1 is out: speed across the board. Exactly faster. Qwen3.8 27B with its drafter +66% on an M5 Ultra, +28% on an M4 Max, +37% on an M1 Pro. Qwen Flash Next: +11% decode on M4 Max, +51% prefill on M5 Ultra. (up to ~5400 PP!!!!) Gemma 4, Qwen, LFM2.5, Spark, Bonsai 2 all faster, output byte-identical to 26.9.6 (across 18 models). New: * GGUF models run on our own MLX engine, written in Zig and Metal, instead of the llama.cpp fallback. (experimental) * 🍣 Sushi-format Flash Next packs run directly. * Qwen-Image 2.1 edits pictures from instructions. * Drafting is on by default for every model that supports it. Huge thanks to everyone who filed, tested and fixed this one: @STRML_, @sbusso, @wutang_superfan , lborloz, LXD-8 @alinselea zeeshanhaque21 @Lojza3D, h9q2cyxvgm-ui @kennethrdegraff, @CowboyCoderHQ , codysk, @Beamsters1, @CerebralCoding_, @AjAbsaki , and more.. THANKS ! Please comment if I did not include your name, for visibility, I know people mostly by their GH handle. Special thanks to @ashxhart and @Spangler3000 for pushing ! Benchmarked and tested across 5 machines: htmlhost.jax.workers.dev/ren… GH Release Download + Changelog: github.com/ddalcu/mlx-serve/… Website: mlxserve.com
48
42
13
393
31,848
Benjamin Marte retweeted
TensorFold V0.3.6.3 is live, with more improvements already in the pipeline. Massive thanks to @ViC305 for the banger of a PR on EXL3, @devopsfortunato for faster CUDA prefill, @AiMan_993 for safer image input, @MiaAI_lab for CUDA tool fixes, @codengod for Gemma support, @ivanfioravanti for helping track down stalls, and everyone contributing code, testing and filing bug reports. Run "tensorfold update" in cli. tensorfold.dev
18
20
3
190
10,840