Pinned Tweet
I am going to Builders & Brews: Tokyo AI Meetup 🗼☕️💻. Join me! luma.com/qa0c59fy via @LumaHQ
2
67
Tim retweeted
Stuck at a machine? Send a photo. Get the next step.
AccessPoint guides older, blind and migrant users through ticket machines and kiosks, in their language. It stops wrong totals and never needs your PIN.
Built on @AIatMeta Muse Spark and SAM 3.1.
accesspoint-bot.aljabrialam.…
you'll know more about inference than 90% of people if you fully understand this article
this is only the first resource in the ai performance engineering repo btw.
imagine the ball knowledge in the other ones
we launched the most comprehensive ai performance engineering repo in the world last week
now we'll be posting every single resource
this is Wafer's ai performance engineering series
save this as your starting point. links in thread 🧵
part 1: "All About Transformer Inference" from How To Scale Your Model.
- the authors cover the computations, memory traffic, and serving decisions behind transformer inference:
- arithmetic intensity of linear layers and attention across prefill and decode.
- kv cache sizing by layer count, kv heads, head dimension, sequence length, and precision.
- the compute/hbm bandwidth crossover and how batch size and quantization shift it.
- decode latency and throughput bounds from parameter bytes, kv bytes, and hardware bandwidth.
- weight reuse through batching and diminishing throughput gains as kv traffic grows.
- gqa, kv quantization, and PagedAttention, including the memory costs each addresses.
- model sharding, kv placement, and collective communication overhead.
- continuous batching, prefix caching, and disaggregated prefill/decode.
the worked problems connect model dimensions to deployment decisions like memory capacity, workload distribution, and expected performance under the stated assumptions.
Tim retweeted
【9/9ハッカソン開催のおしらせ】
CODAS×Tavily×Nebius共同主催!
AIプロジェクトに挑戦しませんか?
経験・チーム・事前のアイデアは不要。
🗓 9/9(水) 16:00〜21:00
📍東京23区内(詳細は参加者にご連絡)
🎟無料
申し込みはこちら👇
luma.com/qa0c59fy
Let’s go Tokyo!!
[Hackathon on Sept 9]
Co-hosted by CODAS × Tavily × Nebius!
Ready to take on an AI project?
No experience, no team, no idea needed.
🗓 Wed, Sept 9 | 4:00–9:00 PM
📍 Tokyo (details shared with registrants)
🎟 Free Register here 👇
luma.com/qa0c59fy
1
83
Organising this btw, let’s go 🇯🇵! Please register if you happen to be in Tokyo, or share to Tokyo friends! @nebiusai @tavilyai @SwarmWork @akindo_io
[Hackathon on Sept 9]
Co-hosted by CODAS × Tavily × Nebius!
Ready to take on an AI project?
No experience, no team, no idea needed.
🗓 Wed, Sept 9 | 4:00–9:00 PM
📍 Tokyo (details shared with registrants)
🎟 Free Register here 👇
luma.com/qa0c59fy
2
107
Congratulations to the Featherless AI team on their $20M Series A. I’m impressed at the infra they are building to make AI accessible and for walking the talk - working with open source builders and researchers for under-resourced languages !
Today we're announcing our $20 M Series A, co-led by @AMDVentures and @AirbusVentures, with @BMWiVentures, @Kickstartph, @PanacheVC, and @Wavemaker_VC.
We're building the neutral layer for open models — so the future doesn't run on a few labs, a few chips, or a few clouds.
THE FUTURE CAN’T BE BLOCKED.
Any Model. Anywhere. Anyone. Yours.
2
83
Tim retweeted
There are still some spots in the room, but they are running low fast!
starting soon, in hall 4 #3
see you all there and follow that thread :)
Tim retweeted
It’s been ages since I shipped code myself—I felt like a COBOL dev lost in the React age. Then vibe coding hit, and I had to try it firsthand. Welcome to ImageBattle ⚔️
📣I'm hosting the @DeepLearningAI Pie and AI London event on the 24th of March 🤖
Slots are limited, sign up to join the waitlist: bit.ly/aihive-meetup
@Chi_Wang_ @ag2oss @IDEALondon @wilbe_science @WilbeLAB
Tim retweeted
Our November Papers of the Month is now live.
This edition covers 4 papers, on "super weights", context-parallelism, scaling laws for precision and critical batch sizes. We provide our summaries and analysis of each. (🧵1/n)
graphcore-research.github.io…
Tim retweeted
1/ Can AI scaling continue through 2030?
We examine whether constraints on power, chip manufacturing, training data, or data center latencies might hinder AI growth. Our analysis suggests that AI scaling can likely continue its current trend through 2030.
Excited for the next part of this journey!
Exciting news: Graphcore joins @SoftBank_Group to build next generation of AI compute. graphcore.ai/posts/graphcore…
1
108
Tim retweeted
Andrej Karpathy talking about -
- rarity of talents in distributed computing with GPUs.
- the opportunities in computer architecture because von Neumann architecture may not be optimal
- progress in precision
- sparsity in neural networks
Tim retweeted
The recording of our talk at @letsunifyai's reading group is now available: youtube.com/watch?v=xq_8dgQF…
Thanks to Unify for having us. It was an excellent session with some great discussion
Really excited to announce @luka_ribar and @savelichic from @graphcoreai as our guests for today's live reading group! 😊They will be presenting "SparQ Attention: Bandwidth-Efficient LLM Inference", which is a technique for increasing the inference throughput of LLMs by reducing the memory bandwidth requirements within the attention blocks 🔥
If you'd like to learn more, then join our meeting room today at 7:00 am PT, 10:00 am ET, 2:00 pm GMT, 3:00 pm CET, 7:30 pm IST ✨ Link below ⬇️
All are welcome, from beginners 🧑💻 to experts 🥷
See you there ✌
Meeting: meet.google.com/qsb-phmd-yxi
Livetream: youtube.com/live/xq_8dgQF5Dk…
Paper: arxiv.org/abs/2312.04985
GitHub: github.com/graphcore-researc…
Tim retweeted
Bringing Large Language Model power to the people! Pienso + Graphcore scoop Best Innovation in NLP at the CogX Awards. #CogXAwards23 #CogX_Festival
hubs.la/Q0220RXF0
Best Innovation in NLP for @graphcoreai & @pienso_ml for enabling low/nocode interactive AI and handling low-latency finetuning and inference in production! Kudos to the teams in both companies across the entire tech stack that made this possible! @CogX_Festival #CogXFestival23
Attended an interesting panel on Industry Transformation in the Creative Industries, panellist are coming from music, digital media, and architecture. Some key insights I found interesting are as follows:
🧵 1/N
1
96
-The upside in GenAI is to democratise the creative tools and production of creative output, this could mean volume of low-quality output (e.g. a landfill of digital image), but it elevated the art form in general. Let’s not underestimate the audience’s ability to appreciate.
3/N
1
44
⁃There’s excitement in the ability of AI to introduce efficiencies and aid in the production process.
⁃There’s concern on lack of governance but also in governance stepping in and hampering advancement.
4/4
#CogXFestival23
66



























