LDVP is a deep-tech venture capital fund. We chose to name our firm LDV Partners in honor of Leonardo Da Vinci.
Palo Alto
Joined June 2026
- Tweets12
- Following234
- Followers14
- Likes62
LDV Partners retweeted
So happy @kevin_flansburg will be joining us at the first SGLang Summit! We’ve been working closely with the Cloudflare team for a long time, and they’ve been such a great team to work with. Looking forward to hearing more about their work and catching up in SF! 😃
Excited to be speaking at the first SGLang Summit (@sgl_project), Nov 12–13 at Fort Mason in SF!
My team (@areneau @KastanDay ) and I are looking forward to presenting our work on system one models (Clef and Clairic), and sharing how we are maximizing KV cache hit rates using a multi-deployment Mooncake cluster with HiCache. Come say hi if you're there!
LDV Partners retweeted
I’ve been working towards AGI my whole life, and as we enter this pivotal moment, I’m stepping into a new role as Chair of Google DeepMind & Chief Scientist of Alphabet. This will allow me to focus on long-term strategy, and accelerating scientific breakthroughs, including leaning into my work at Isomorphic to help cure disease.
I’m excited that @koraykv will be stepping up to lead GDM as SVP, alongside @joshwoodward and our exec team. I could not be more excited and confident about our amazing next chapter! 🚀
blog.google/company-news/ins…
LDV Partners retweeted
RadixArk's mission is to make frontier AI infrastructure open and accessible to every builder. SGLang embodies that, and we're excited to work with Google Cloud @googlecloud @googledevs to bring its performance and flexibility to the TPU ecosystem.
RadixArk and Google Cloud are joining forces with the SGLang community to make TPU a drop-in, cost-efficient path to frontier inference.
SGL-JAX already serves the major open model families on the latest TPU generations: Gemma, Qwen, DeepSeek, GLM, Kimi, Ling, MiniMax, MiMo, Grok, plus Wan and Flux for video and image generation.
With this partnership, TPU unlocks SGLang's production feature set: 5D parallelism, Radix Cache, HiCache, quantization, speculative decoding, on TPU Pallas kernels co-built by Google, RadixArk, and the SGLang community. SGL-torchtpu opens a PyTorch-native path to all of it.
SGLang on TPU comes with the same API and the same features, so for teams already running it, getting started is simple. Going forward, TPU support for new open models will land alongside the rest, and that carries forward to each new TPU generation.
Full announcement and SGL-JAX repo 👇
LDV Partners retweeted
RadixArk has proudly co-signed this letter.
Our mission is to make frontier AI infrastructure open and accessible to everyone. We power the open-weight ecosystem with SGLang for serving and Miles for training.
Excited to build the future of AI together with the community!
For my first post, I’m sharing a letter @NVIDIA signed on why open models matter.
AI will transform every industry, power every company, and be built by every country.
Open models strengthen safety and cybersecurity, accelerate innovation and diffusion, and enable sovereignty.
The world needs both frontier closed models and frontier open models.
images.nvidia.com/pdf/Open-W…
LDV Partners retweeted
Jensen's first post and it is a letter on why open models matter.
Working in open source inference, I see the diffusion argument play out daily. Teams across finance, telecom and automotive are moving open models into production on NVIDIA hardware right now, many of them in places a closed API could never reach for sovereignty or compliance reasons.
The world needs both frontier closed and frontier open models. Could not agree more. Welcome to X @JensenHuang and thank you for putting the weight of NVIDIA behind this!
For my first post, I’m sharing a letter @NVIDIA signed on why open models matter.
AI will transform every industry, power every company, and be built by every country.
Open models strengthen safety and cybersecurity, accelerate innovation and diffusion, and enable sovereignty.
The world needs both frontier closed models and frontier open models.
images.nvidia.com/pdf/Open-W…
LDV Partners retweeted
Quantum computing has always felt “10 years away.”
Now, experts are saying it’s closer than you think. Here are 10 terms to help you understand what’s coming next: msft.it/6004QfePO
LDV Partners retweeted
Big fan of TML’s mission to build AI that extends human will and judgment, and reimagine how humans and AI collaborate.
With Inkling, released with full weights and built for customization, we see a great future of US open models. Congrats, @thinkymachines!
Today, we are introducing Inkling.
Inkling reasons efficiently across text, image, and audio modalities. We are making the full weights available.
thinkingmachines.ai/news/int…
Available today for fine-tuning on Tinker. Play with it in the Inkling Playground. 🧵
LDV Partners retweeted
🎉 SGLang v0.5.15 is out!
We spent this cycle tuning GLM-5.2 NVFP4 for production serving, now hitting 500+ tok/s/user on 8x B300 and 450 on 4x GB300 (bs=1).
We will put commands to run this at the thread below, and full technical details and instructions on a blog very soon 🫡
And we have some newly supported models: Hunyuan 3 (Hy3), Hierarchical Reasoning Model (HRM-Text), NVIDIA LocateAnything-3B, Baidu Unlimited-OCR, JoyEcho, and Qwen3.6.
Here are highlights for this release:
- Breakable CUDA Graph is now the default capture path
- Native web search built in, powered by @ExaAILabs
- Decode context parallelism for MLA models, including DeepSeek V3
- FlashInfer all-to-all for routed MoE
- DeepSeek-V4: FlashMLA sparse prefill now on by default (>10% throughput on long context), plus a non-paged indexer for long-context prefill (>5% e2e)
We welcomed 43 new contributors, and thanks again for our amazing partners and model makers: @NVIDIAAI @AMD @intel @Zai_org @TencentHunyuan @Alibaba_Qwen @deepseek_ai @Sapient_Int
Now. MAX LOAD! MAX OUTPUT! 🚀
LDV Partners retweeted
Congrats @humansand on native NVFP4 training recipe, landing in Miles!
LDV Partners retweeted
Miles is now featured on the PyTorch Foundation blog.
As models grow, shift from dense to MoE, and span more specialized hardware, RL post-training is no longer just about the algorithm. It is a distributed systems problem. Miles is our open-source RL training framework, built for exactly that.
It comprises four systems behind a small, pluggable trainer: SGLang (@sgl_project) for rollout, Megatron-LM (@NVIDIAAI) for training, Ray (@raydistributed) for orchestration, and PyTorch (@PyTorch) as the common layer for models and numerics.
Out of the box, you also get MoE-aware rollout/training alignment, a unified BF16/FP8/MXFP8/NVFP4/INT4-QAT pipeline, fast NCCL/RDMA weight sync, fault tolerance, and ready-to-run recipes for frontier models like DeepSeek V4, GLM 5.2, Qwen3.6, Kimi K2.6, and Nemotron 3 Ultra.
Our goal is simple: make frontier-scale LLM RL easier to reproduce, extend, and operate.
Thank you, PyTorch Foundation, and everyone who got Miles here, especially the legendary @slime_framework team!
Built on PyTorch, Ray, SGLang, and NVIDIA Megatron-LM, Miles is an open source framework from RadixArk for large-scale LLM reinforcement learning post-training.
Miles uses PyTorch for models, numerics, profiling, and extensibility; Ray for orchestration; SGLang for rollout generation; and Megatron-LM for distributed training.
The framework supports asynchronous rollout and training, NCCL/RDMA weight synchronization, MoE-aware rollout/training alignment, low-precision recipes, LoRA, fault tolerance, observability, and extension points for custom algorithms and model architectures.
🔗 Read more in our latest blog from the Miles Team: pytorch.org/blog/miles-a-pyt…
LDV Partners retweeted
Day-0 support is the starting point, not the finish line. The SGLang community has always been turning frontier models into fast, production-ready serving systems as quickly as possible.
In 2 months, folks achieve 5X throughout speedup in Nvidia GB300.
Proud of the @sgl_project and @NVIDIAAI teams pushing this forward.
While SGLang provided Day-0 support for DeepSeek-V4, the collaboration between the @lmsysorg and @NVIDIAAI engineering teams has taken its production performance to the next level.
According to the public SemiAnalysis InferenceX dashboard, the GB300 disaggregated lane (DeepSeek-V4 Pro, FP4, 8K/1K) saw a 5x throughput increase—surging from ~2,200 to ~11,200 tok/s/GPU at identical interactivity levels. These updates sustain high throughput much deeper into target interactivity ranges most deployments target, while also driving a 2.9x lift on the Blackwell Ultra aggregated lane.
Find the full technical breakdown in the comments below:
