@IQReactorAI

AI in Public 🛠️ | ex @SentinelOne 1st SRE Go Team & @Comcast VIPER Startup | Building : Strix Halo / R9700 K8S Cluster for Compound AI Systems

 🏔 DENVER 🏔
iqreactor.com/links
Joined September 2016
  • Tweets1.9K
  • Following306
  • Followers1.2K
106Media
Ryan Wright retweeted
Carlo
@ItalianclownzOct 24
Hearing the creator of Mario, Shigeru Miyamoto, say that the Typerwriter to Word Processor is the equilivent of Generative AI and is part of technological human evolution is probably one of the best statements of 2026. It will not make human creativity obsolete Human intention and individual visions will become even more important
1
1
15
827
Ryan Wright retweeted
sleepy
@sleepycorp_Aug 30
DIY tower 3d printed ✅ Awaiting fans from Amazon now Dual Asus GX10 setup is getting nicer every day between new recipes from @MiaAI_lab and creators in the 3d printing space haha
9
4
50
4,082
Ryan Wright retweeted
Loktar 🇺🇸
@loktar00Aug 29
As I'm waiting for three models to generate some content I was just remembering we used to have models that didn't think and had 2048 context windows. We've come so far.
7
2
46
2,109
Ryan Wright retweeted
Caleb Gross
@noperatorAug 8
use LLMs not to offload understanding, but deepen understanding
2
2
19
3,102
154
770
74
14,844
565,764
Ryan Wright retweeted
Trevor Wood
@trevorwood222Aug 4
Replying to @IQReactorAI
This was done on a phone. 8B getting 20-30 tok/s xitter.cf/trevorwood222/status/2…
Trevor Wood
@trevorwood222Aug 3
lfm 2.5 8b a1b abliterated Device: Red Magic 9s Pro 16gb Harness: PocketAI Hub Android App OS: Android 16 Base Q4: 21.63 tok/s, 1,770 tkns, 8 GB peak RAM Abilterated Q4: 21.67 tok/s, 1,327 tkns, 8 GB peak RAM
1
1
1
87
Ryan Wright retweeted
CNX Software
@cnxsoftJul 24
New @AMDembedded Ryzen AI Embedded X100 CPU family, COM-HPC module, and developer platform. cnx-software.com/2026/07/24/… The 55W Ryzen AI Embedded X100 family offers up to 16 Zen 5 cores, an RDNA 3.5 iGPU, and up to 50 TOPS NPU. The company also provides the Kria AI SOM COM-HPC module featuring the Ryzen AI Embedded X199 CPU and up to 128LPDDR5, and the Kria AI Robotics developer platform with 10GbE networking, FAKRA camera connector, an Oculink connector, and more. The new family supports Linux, the AMD ROCm software stack, Xen Hypervisor, and standard AI frameworks like PyTorch, ONNX, and TensorFlow. It's designed to compete against Intel Core Ultra Series 3 and NVIDIA Jetson Thor T5000 solutions.
1
22
6
134
12,413
Ryan Wright retweeted
Sandro
@pupposandroJul 23
AMD Gorgon Halo Cluster preview. 2x 192GB unified memory, same Strix Halo memory bandwidth.
32
25
4
376
23,714
Ryan Wright retweeted
Ahmad
@TheAhmadOsmanJul 22
This could actually be massive CUDA compiler for different hardware platforms would be a game changer for Inference Optimization if implemented up to spec
Eric Hartford
@QuixiAIJul 22
Y'all... I don't think you get it yet. *anything* runs on *anything* let this sink in.
16
30
304
25,284
Ported our 8x GB10 cluster to the v16 unified base + b12x sparse indexer. All measured, TP8. prefill pp2048: 880 -> 1,101 tok/s @4k depth (+28% vs our old stack) depth prefill: 1052 / 940 / 838 at 77K / 231K / 395K (was 653/441/319) single-stream decode: 42.1 tok/s median 860K max_len serving, KV pool 903K tokens. needle recovered at 845K, still decoding 10.9 tok/s at that depth. Two traps if you're on this base: cold inductor cache makes first-boot lazy compiles (~53s each) look exactly like a shm-broadcast hang. warm restart -> 3s, gone. and available KV drifts 47->41 GiB over repeated boots from GPU memory fragmentation. full node reboot resets it. Thanks @anuntrapid_auto your repo + patch pointer were the starting point. GLM-5.2 Int4-Int8Mix, fp8 sparse-MLA KV, MTP. xitter.cf/anuntrapid_auto/status…
llmwish.open
@ciprianvegJul 14
GLM-5.2-Int4-Int8 on 8× GB10: ~1,200 t/s prefill, 33–54 t/s avg decode (generic - coding/structured) and memory remaining to run also a Mimo 2.5 in parallel for image/audio input, both tp 8. forums.developer.nvidia.com/… github.com/ciprianveg/gb10-g…
11
26
8
149
44,580
New benchmark just dropped!
Alex
@The_AlexJul 15
Tell Codex to open Microsoft Paint and try to draw you
12
31
2
542
39,466
Ryan Wright retweeted
Chaofan Shou
@shouccccMar 31
Claude code source code has been leaked via a map file in their npm registry! Code: pub-aea8527898604c1bbb12468b…
3,261
7,304
5,313
47,686
35,907,677
Ryan Wright retweeted
Min Choi
@minchoi28 Aug 2025
Google Nano Banana just broke the Internet yesterday People are getting crazy creative with it. 10 wild examples:
121
525
37
8,526
1,512,298
365 days of UNLIMITED FREE Nano Banana. Only for Higgsfield Pro, Ultimate & Creator users. Limited-time offer until the end of this week. You read that right. Unlimited & Free for one WHOLE year. You've got 4 days to secure the offer. A HUGE Banana, indeed. Go wild. Retweet to get the offer link in DMs 📨
143
531
175
1,461
1,307,323
Ryan Wright retweeted
Kyle⚡@elyktrix18 Jan 2022
You can absolutely learn JavaScript in a week. Every week.
66
112
15
1,242
Spend 80% of your time with people who help you generate 80% of your desired results.
1
1
9
Ryan Wright retweeted
Dr. Tech
@doctechmd30 Dec 2021
Dear developers, good night! 🌚🦸‍♂️ People deserve to know the truth about programmers
2
6
2
62
✅ How to output QR codes in your app? Its simple. 👇Attached are 2 code examples from Python and JavaScript (Photos) 🧠 An interesting fact about the data they hold : Numeric. 7,089 chars Alphanumeric 4,296 chars That's actually alot the data !
Example of generating QR codes in Python

ALT Example of generating QR codes in Python

Example of creating QR codes in Javascript

ALT Example of creating QR codes in Javascript

This quoted post is unavailable.
Try viewing anyway · View on X
2
3
=> Free hosting for static sites <= ✅ GitHub Pages ✅ Netifly ✅ Firebase ✅ Amazon S3 ✅ Zeit ✅ Forge
1
2
Ryan Wright retweeted
Bioshock@bioshock_hk17 Dec 2021
Raspberry_Pi: Way back in the 1980s, back before the World Wide Web, if you lived in France you could send messages, book train tickets, and do all kinds of other useful things via Minitel. Gautier Serodon gives a vintage terminal a new lease of life wit…
A vintage Minitel terminal running Raspberry Pi OS

ALT A vintage Minitel terminal running Raspberry Pi OS

1
6
1
12
Load more