Select difficulty 👁️🗨️
United States
Joined December 2017
- Tweets392
- Following559
- Followers729
- Likes4.4K
In the future you’ll grow up with model by your side - one that knows your strengths, understands your weaknesses, and helps you become your best self. The future is bright. One day, we’ll look back and marvel at how far the human condition has advanced.
"Missing after_thread_sync"
We’re releasing our Blackwell kernel verifier, building on our recently released work on "AI as a Compiler." In this approach, an LLM generates optimized PTX for a given workload, and our verifier checks it for correctness. By skipping intermediate compiler layers, we enable finer-grained control over the hardware at the PTX level, paving the way for more efficient kernels and models.
The verifier checks several properties, including symbolic correctness, race conditions, and synchronization issues.
You can get access here:
Volta Blackwell Verifier: github.com/francois141/ptx_v…
Triton AI Compiler: github.com/francois141/trito…
And here is the paper:
AI as a Compiler: arxiv.org/abs/2609.36800
Alex Zhurkevich retweeted
No one changes the world alone
Introducing Ampersand - the first model built for us. Ampersand is a collaborator trained to help people achieve their individual and collective goals. In multi-month simulations, users collaborating with Ampersand accomplished more than with Fable and other frontier models
Available to everyone through our API
Alex Zhurkevich retweeted
Everyone talks about Palantir inventing the FDE, but I'm pretty sure @nvidia's DevTech org predates that
DevTech is one of the most unique and interesting parts of NVIDIA. The best programmers in the world ready to airdrop into a new customer's code at a moment's notice
Alex Zhurkevich retweeted
GPT-6 Astra is insanely good. But expensive
So we're giving it away for free.
We want everyone to benefit from the frontier of intelligence
platform.xplabs.ai
Alex Zhurkevich retweeted
big release
Today we're launching Miles v0.1, an open-source RL framework for LLMs and multimodal models.
RL training is easy to start and hard to debug. Miles helps you ensure your run is correct, use hardware efficiently, and keep RL running at scale.
Over the past 9 months, 72 contributors have landed 1,326 commits, 85 GPU E2E CI tests, battle-testing Miles on frontier open models like Kimi K3, DeepSeek V4, Qwen 3.8, GLM 5.2, Inkling, MiniMax H3, etc.
Miles powers frontier-model development and production RL workloads at @humansand, @periodiclabs, @modal, @DecagonAI, @Eigent_AI, @nebiusai, @IBM and more, on both @NVIDIAAI and @AIatAMD hardware.
Here is what we built, and why teams picked Miles🧵
CUDA PTX simplicity with Python syntax. Compatible with CuTe if that’s your workflow 🫡
Together with the new primitives API, CuteDSL 4.7.0 now has explicit host-side TMA creation, mirroring CUDA C++'s cuTensorMapEncodeTiled
no more make_tiled_tma_atom and tma_partition 🥳
Thanks @Leik0w0 for sharing with me this
Alex Zhurkevich retweeted
Today we’re officially launching @explabsai on YC.
Companies spend more on AI every month, but none of it becomes an asset they own.
Up to 97% cheaper and 50% higher quality.
Alex Zhurkevich retweeted
🍰 and another CAKE kernel merged into FlashInfer: SM100/SM103 tinygemm2, up to 1.79x faster on large shapes, ~18-23% geomean kernel-time reduction, e2e verified.
translates to up to 7.6% end-to-end serving speedup in SGLang.
github.com/flashinfer-ai/fla…
