We closed our series D at $2.1b. It happened 8 months after our series C, which seems too fast until you consider the facts: 2-years worth of growth in 8 months, virtually 0 customer churn, healthy margins, and QoQ NDR numbers that are considered top-tier YoY. The market demand

Amazing work by @drfeifei and the @worldlabs team to push the boundaries of multimodal AI.

with the lowest time-to-first-token:

A few days ago Kimi K2 Thinking significantly narrowed the capability gap between open and closed LLMs. Today Baseten is the only provider to deliver over 100 tok/sec on this massive 1T-parameter model.

The fastest provider for kimi-k2-thinking <now in Cline>

Cursor 2.0 feels game changing - fast agentic workflows unlock new levels of creativity and productivity. Congrats to the team!

And we have the highest tok/sec using nvidia GPUs:

There's an obsession with tok/sec as *the* metric in LLM inference. But in latency-sensitive use cases the metic that matters more is time-to-first-token: - Code edit use cases have short outputs and overall latency is heavily determined by ttft - Voice AI use cases care about

GLM-4.6 providers overview: we are benchmarking API endpoints offered by Baseten, GMI, Parasail, Novita, Deepinfra GLM-4.6 (Reasoning) from @Zai_org is one of the most intelligent open weights models, with intelligence close to GPT-OSS-120b (high), DeepSeek V3.2 Exp (Reasoning)

We see the massive AWS outage. Baseten web app is down but inference, new deploys, training jobs, and the model management APIs are unaffected.

GLM 4.6 fans! @basetenco just soared to the top as the fastest provider in Artificial Analysis for the model. > a 114 TPS and <0.18s TTFT. > That's 2x faster than the next best option on both metrics. Available now in Cline.

Go team @FactoryAI!

What do Superhuman, Baseten, and Ricky Bobby all have in common? An obsession with speed. If you’re a Superhuman user, you know their email app lives and dies by how fast their users can get through all things email.

If you see a doctor today, chances are they're using OpenEvidence for trustworthy, up-to-date medical information at their fingertips. We're thrilled to support OpenEvidence's mission with the speed (<160 ms latency) and reliability physicians require at the point of care.

@rtfeldman teaching me zed live now!! https://t.co/cRPiwkD0sT

📈 @basetenco users are scaling smarter with us: ✅ 5× throughput on high-traffic endpoints ✅ 50% lower cost per token ✅ Up to 38% lower latency on the largest LLMs Built on NVIDIA Blackwell + TensorRT-LLM + Dynamo on @googlecloud—driving efficiency, speed & adoption at scale.

Next live stream is tomorrow at 10:30AM pst. Sharing some fun announcements, guests joining, and we're giving away a bunch of our shirts!

🚀 Arcana v2 is here. Rime’s next-gen TTS makes voice AI sound truly human. More languages. More realism. More deployment options. 🧵👇

It's important to support newly released open-weight models on day 1. But it's not noteworthy. What's noteworthy is to have the inference optimization muscle to immediately blow the competition out of water on latency and throughput. As measured by OpenRouter:

We're very excited to be an @OpenAI launch partner for GPT OSS. Today's a big day for open models, and we have day 0 support for GPT OSS 120b via our Model APIs: https://t.co/hMLzTy2dek We'll be rolling out more performance optimizations and benchmarks over the coming hours and