appughar retweeted
If you’re at @EurIPSConf this week and looking for your next ML role, ping us. Few of us from @Dropbox including @mobicham @appughar will be there Happy to chat about roles in multimodal, search, and video-AI.
Replying to @Mobius_Labs
Looking for people with experience in : Building and scaling multimodal ML systems, Retrieval, ranking, and creative search problems, Turning prototypes into polished products
🌍 Remote (Germany or Denmark)
Apply here: jobs.dropbox.com/listing/738… (Please use the link, not DMs!)
appughar retweeted
GemLite now supports @vllm_project vllm V1, which brings up to 1.25x faster inference speed vs V0!
github.com/mobiusml/gemlite/…
appughar retweeted
Well optimized Triton kernels can perform very well end-2-end, even competing with highly optimized kernels like Marlin.
appughar retweeted
GemLite now supports bfloat16 on both Hopper and non-Hopper gpus 🫡
github.com/mobiusml/gemlite/…
appughar retweeted
Our re-distilled @deepseek_ai R1 (1.5B) outperforms the original distilled model! Get it at huggingface.co/mobiuslabsgmb…. We’re distilling more models and look forward to releasing them soon!
appughar retweeted
i'm comically impressed that people are coping on deepseek by spewing bizarre conspiracy theories -- despite deepseek open-sourcing and writing some of the most detail oriented papers ever.
read. replicate. compete.
don't be salty, just makes you look incompetent.
appughar retweeted
Heard you like distillation, so we re-distilled the distilled model
Our re-distilled @deepseek_ai R1 (1.5B) outperforms the original distilled model! Get it at huggingface.co/mobiuslabsgmb…. We’re distilling more models and look forward to releasing them soon!
appughar retweeted
It was a real pleasure and a productive experience collaborating with the team at
@PyTorch and the SGlang team at
@lmsysorg to bring gemlite to TorchAO and SGLang. This is an exciting area that requires a lot of work. Let’s push the boundaries of efficient computing!
Scaling Large Language Models (#LLMs) is tough due to resource demands. Quantization helps but often comes with trade-offs in accuracy and performance.
Discover how #PyTorch, Mobius Labs, and SGLang tackled these challenges with Gemlite, TorchAO, and SGLang for efficient low-precision inference.
📖 Read here: hubs.la/Q033z_kw0
appughar retweeted
Releasing a new version of Gemlite github.com/mobiusml/gemlite significantly improved performance on datacenter GPUS (A100/H100) delivering up to 7–8x faster prefill and 3–6x faster decoding compared to PyTorch's tinygemm.
appughar retweeted
Check out our own @mobicham's lecture on building low-bit Triton kernels & the impressive speedups we’re achieving with them, presented at @GPU_MODE channel. Also includes plenty of tips & tricks for performance tuning in Triton & GPU programming.
youtube.com/watch?v=7c3c3bCG…
appughar retweeted
Replying to @mobicham
@mobicham will be talking about low-bit Triton kernels and his experience building github.com/mobiusml/gemlite this Saturday at the GPU mode discord. It’s sure to be a great session if you’re interested in efficiency!
discord.gg/KKkmtQPx?event=12…
appughar retweeted
I really like the newly unveiled statue of Satoshi Nakamoto in #Lugano. Especially the way it honors privacy. Depending on the perspective, it disappears. #PlanBForum #privacy 🇨🇭
appughar retweeted
At Mobius Labs, with our academic roots, we often debate publishing papers. But business pressures and the angst about the prodigal reviewer bent on pressing the reject button, have led us to focus on blogs and open-source work instead. Check some of it at blog.mobiuslabs.com
ICLR 2025 will have another blogpost track!
If you have new intuitions on past work, noticed key implementation details for reproducibility, have insights into the societal implications of AI, or an interesting negative result, consider writing and submitting a blogpost. Accepted entries will have the opportunity to be presented as a poster at ICLR2025 (in Singapore)! Deadline: Before Nov. 16 2024
More info: iclr-blogposts.github.io/202…
appughar retweeted
We’re thrilled to announce that our team at Mobius Labs has won Runner-Up in the SoccerNet Game State Reconstruction Challenge, in collaboration with @IRI_robotics 🏆⚽
appughar retweeted
This! One of the pleasures of open source is building together. You learn, iterate, & share ideas fast. Startups stuck in closed environments only reach local optima at best. Open source expands your perspective, connects you with great people, & delivers quality with quantity.
appughar retweeted
We benchmarked the new OpenAI's whisper-turbo model using long-form audio (5+ mins) with faster-whisper ( github.com/SYSTRAN/faster-wh…) backend. Results ( as-expected): It achieves WER comparable to large models and excels in speed among multilingual models.
appughar retweeted
We added the support for safetensors for HQQ models in @huggingface . Thanks a lot for the good folks there who helped us with this PR. Hoping to see a lot of HQQ quants in the hub!
appughar retweeted
CUDA MODE has grown up! -> GPU MODE
From the beginning, @neurosp1ke and I wanted a community that was cozy, inclusive and fun where you could make friends and learn how to make your code faster
Anyone who shares our ideals is welcome. We’re announcing we’re changing the name of our community to “GPU MODE”.

















