DigitalOcean

Another Deploy in the books. 📖 A special thanks to our attendees, sponsors, speakers, & their companies. 🤝 @nvidia @arcee_ai @VAST_Data, @AMD @weaviate_io @OpenRouter @MongoDB @inferact @hippocraticai, @BessemerVP @character_ai Premise VC @Workato Andi @500GlobalVC @ISMG_News @LawVo @higgsfield Probably AI @AntlerGlobal

@piq9117 Hi there, we’re sorry about the error you’re facing. Could you please share the ticket number if you’ve already created one for this, or DM us your email address so we can create a ticket on your behalf and have our team look into this? We’re here to help!

@digitalocean what is happening here?

@NVIDIADC @vllm_project @inferact So grateful for the partnerships and thanks for amplifying the excitement, @NVIDIADC! 🌊

Huge milestone for @digitalocean — and a great example of close collaboration with hardware-software co-design at every layer. NVIDIA Blackwell Ultra + NVFP4 + optimized @vllm_project stack for speculative decoding with @inferact and the open source community → #1 performance on Artificial Analysis for leading frontier models. More to come as this ecosystem scales.

@nininitom Here's your ticket# 12127887, please keep an eye on your inbox for any updates associated to the issue you're experiencing.

@digitalocean The email is tom.k.kipkurui@gmail.com

Hey @DOStatus , I want to create a droplet but I get this instead

@nininitom Appreciate you for providing us the email address. We'll go ahead a create a proactive ticket for you and route this to relevant team, you can expect a reply sooner than anticipated.

Deploy today. 🚀 do.co/4tNVIgY

DeepSeek-V4-Pro is available on DigitalOcean. 🌊 Bring planning-first intelligence to your apps with @deepseek_ai's latest frontier model. ✅ 1M token context window ✅ Built for agentic reasoning and multi-setep tasks ✅Run alongside your apps & data on DigitialOcean ✅Pay only for what you use. Scale your AI systems without the infra headache.

@loftwah Thank you for this. We have sent a DM, requesting for more information. Thank you for your patience in this matter.

@digitalocean ref:!00Df2018t5m.!500QP01RKKXE:ref is the ticket number.

Hey @digitalocean any reason I would be getting billed for an account I deactivated months ago now?

@stecal12 We do applogize for this. We have sent you a DM requesting your account details, so we can make a ticket for the Support Team on your behalf. Thank you for your patience in this matter.

@digitalocean Thanks for punting the problem back to the customer. How about I just move to Vultr instead? Isn't the problem already described in enough detail for you to take ownership and address the issue? Using a gmail account, the sending issue is on you.

@digitalocean Email for 2FA isn't the best idea in the world, if that is your only option for this. Can't do passcode or SMS, or just let me use my password (my security, my choice). 10 minutes on and still waiting for the code.

We just set the bar for inference performance. Fastest DeepSeek, MiniMax & Qwen on Blackwell Ultra. #1 output speed. Sub-1s latency on @digitalocean's AI-Native Cloud Platform as validated by @ArtificialAnlys 230 tok/s on DeepSeek V3.2 which is 3.9x faster than AWS Bedrock and Sub-1s TTFT. This is what we unveiled at Deploy earlier this week: 👉 An AI-Native Cloud built for the Inference Era 👉 A unified Inference Engine — speed, cost, simplicity Red the article below to understand how we engineered this with techniques inlcuding Tensor Parallelism, Kernel Fusion, Programmatic Dependent Launch, Speculative Decoding and Multi-Token Prediction (MTP) including working closely with the creators of vLLM at @inferact digitalocean.com/blog/how-we-bu…

@qencode We're so grateful to have spent the day together, @qencode and thanks for sharing the excitement! 🌊

We spent the day at @digitalocean Deploy in SF, where DO just launched 15 new products and several updates to expand their cloud inference infrastructure. Congratulations to all the teams involved in putting on this phenomenal event! #DODeploy #DigitalOceanDeploy26

@Workato @nvidia @arcee_ai @VAST_Data @AMD @weaviate_io @OpenRouter @MongoDB @inferact @hippocraticai @BessemerVP @character_ai @500GlobalVC @ISMG_News It was such a pleasure to have the @Workato team there. Excited for what more we'll be able to accomplish together this year! 🙌

@digitalocean @nvidia @arcee_ai @VAST_Data @AMD @weaviate_io @OpenRouter @MongoDB @inferact @hippocraticai @BessemerVP @character_ai @500GlobalVC @ISMG_News Thank you for the opportunity to be a part of Deploy!

Another Deploy in the books. 📖 A special thanks to our attendees, sponsors, speakers, & their companies. 🤝 @nvidia @arcee_ai @VAST_Data, @AMD @weaviate_io @OpenRouter @MongoDB @inferact @hippocraticai, @BessemerVP @character_ai Premise VC @Workato Andi @500GlobalVC @ISMG_News @LawVo @higgsfield Probably AI @AntlerGlobal

@simon_mo_ @vllm_project @nvidia @inferact Thanks for sharing the excitement, @simon_mo_ ! We're proud to build with the @vllm_project team and keep pushing the numbers together. 🤝

Top speed, backed by open source @vllm_project, on @nvidia Blackwell Ultra, with the @inferact team, on @digitalocean inference.

@vllm_project @NVIDIAAI Couldn't have done it without the @vllm_project team. Co-designing every layer of the stack with partners like you and @NVIDIAAI is exactly how we get here. 🙌

⚡Congrats to @digitalocean for leaderboard topping fastest speed! @vllm_project open source community is excited to showcase the capabilities of our engine on @NVIDIAAI Blackwell Ultra and powering production AI together!

@vllm_project This is what open collaboration looks like in practice. Optimizations built together, landed back in open-source vLLM. Proud to build with the @vllm_project team and @nvidia. The numbers speak for themselves. 💯🤝🌊

🏆 vLLM powers the fastest inference on NVIDIA Blackwell Ultra on Artificial Analysis. On @digitalocean's Serverless Inference, powered by vLLM on NVIDIA HGX B300: 🥇 AA #1 output speed for DeepSeek V3.2 (230 tok/s, 0.96s TTFT) and Qwen 3.5 397B 🔧 MiniMax-M2.5: 23% TPOT gain via an EAGLE3 draft model trained on TorchSpec Co-design highlights: - NVFP4 quantization on Blackwell Ultra - EAGLE3 + MTP speculative decoding - Per-model kernel fusion Thanks to @digitalocean, @nvidia, and @inferact for the collaboration. Optimizations land back in open-source vLLM. 🔗 digitalocean.com/blog/how-we-bu…

Learn more. 👇 do.co/3QGOTPz

Among the fastest DeepSeek V3.2, MiniMax-M2.5, and Qwen 3.5 397B inference in the market, per Artificial Analysis benchmarks (April 2026). ⚡️🤖 Sub-1-second TTFT. 230 tokens per second. Co-designed every layer of the stack with @Inferact, performance optimized @vllm_project, all on @NVIDIA HGX B300. Live on DigitalOcean Serverless Inference now. Full breakdown in the comments. ⬇️