Fireworks AI

HumanX is just a week away! AI is changing who builds software. Real-time code generation, natural language interfaces, and agentic workflows are opening up development to people who never would have written a line of code before. What does that mean for what gets built, and who builds it? Our Co-founder & CEO Lin Qiao will dig into these questions with Replit’s Amjad Masad,  Imbue’s Kanjun Qiu, and Business Insider's Charles Rollet. - HumanX main stage - Wednesday, April 8th. - 11:30am Right after at 12:00pm, Lin is doing an open Q&A in Room 210. First-come, first-served - bring your questions. https://lnkd.in/gumPtwqU We’d love to see you there! Register now: https://lnkd.in/dTBzSzZ5

This Thursday, we’re teaming up with Exa and Vercel for a night focused on how modern AI apps actually get built. Technical talks, live demos, and a real walkthrough of how fast inference (Fireworks), search (Exa), and application infrastructure (Vercel) come together in production. We'll also sit down for a panel with leaders from each company on where software is headed when you're building for AI, not just humans. → Fireworks: Rob Ferguson - VP of Technology & Strategy → Exa: Jeeyoung Kim - Head of Engineering → Vercel: Andrew Qu, Chief of Software If you’re an engineer or builder working on AI systems, this one’s for you. Click here to register: https://luma.com/nq14wi4o → 6pm - 9pm → Mission Rock, San Francisco

Fireworks is sponsoring the YC W26 Afterparty! Join us for a night with fellow Y Combinator founders and builders. Free drinks, live DJ, and good conversations once the pressure’s off. Hope to see you there! Register: https://lnkd.in/g8vVYun4

𝗣𝗮𝗿𝘁𝗻𝗲𝗿 𝗦𝗽𝗼𝘁𝗹𝗶𝗴𝗵𝘁: 𝗙𝗶𝗿𝗲𝘄𝗼𝗿𝗸𝘀 𝗔𝗜 𝘅 𝗖𝘂𝗿𝘀𝗼𝗿 Cursor's Composer 2 launch last week was a reminder that the frontier is no longer just about building bigger & better base models. Increasingly, it is about what happens next: continued pre-training, reinforcement learning (RL), and the infrastructure required to make those gains real in production. There is transition happening in AI right now:  𝗥𝗟 𝗶𝘀 𝗺𝗼𝘃𝗶𝗻𝗴 𝗳𝗿𝗼𝗺 𝗲𝘅𝗽𝗲𝗿𝗶𝗺𝗲𝗻𝘁𝗮𝘁𝗶𝗼𝗻 𝗶𝗻𝘁𝗼 𝗽𝗿𝗼𝗱𝘂𝗰𝘁𝗶𝗼𝗻 Composer 2 is the best example to-date. It is now a frontier-level coding model, the product of a stronger base model (Kimi K2.5) combined with reinforcement learning on long-horizon coding tasks. Fireworks powered the RL inference behind Composer 2, helping keep large MoE rollout fleets stable and efficient so Cursor can deliver faster, more accurate AI-powered coding. For senior AI leaders, this is the signal worth paying attention to: The bottleneck is no longer just access to the newest models. It is the ability to tune them on real application data, improve them against real workflows, and serve them efficiently and reliably at scale. The next wave of frontier AI will not be won by model access alone. It will be won by the teams that can turn post-training and RL into durable product performance. Want to get into the details? Take double click below. 👉Cursor’s full Composer 2 report: https://lnkd.in/gKDjtnJ8 👉Fireworks blog: https://lnkd.in/gkYPjxYq

San Francisco! You have a few hours left to register for the Fireworks on Microsoft Foundry Launch Event! A live deep dive into Fireworks on Microsoft Foundry, showing how developers are deploying open models with high-performance inference directly inside Azure. Live demo. Technical walkthrough. Good food, drinks, and people. Hear from: → Rob Ferguson, VP of Technology & Strategy, Fireworks AI → Pranay Bhatia, Director of Product, Fireworks AI → Dave Voutila, Principal Product Manager, Microsoft Azure AI Register here: https://msft.it/6044QlJ7C Learn more about Microsoft Foundry + Fireworks: https://lnkd.in/dp3D-RFb → March 26, 5:30 PM → San Francisco → Fireworks + Microsoft

Proud to have a team of rockstars like Roberto solving hard problems for our partners.

From the pregame show to the afterparty: Fireworks is bringing the intelligence behind the world's best AI companies. Come to Fireworks.ai to learn. #GTC26 #Nvidia

Cursor's Composer 2 beats Opus on TerminalBench at a fraction of the cost. Big results like this don’t happen by accident. Fireworks is proud to have powered the inference and RL scaling behind Composer 2.

Fireworks made a big splash at NVIDIA #GTC Can't wait to see you all next year!! #GTC #GTC26 #NvidiaGTC

Challenge: Autonomous AI agents are the future, but building them to be truly safe, secure, and efficient at scale remains a major hurdle. Solution: NVIDIA NemoClaw + Fireworks We’re excited to collaborate with NVIDIA on NVIDIA NemoClaw, an open source stack that simplifies running OpenClaw always-on assistants, more safely, with a single command. *Why it Matters for Developers* Simplifies Security: It safely runs always-on OpenClaw assistants with a single command. Secure Runtime: Installs the NVIDIA OpenShell runtime, giving your agents and open models the access they need while enforcing policy-based security, network, and privacy guardrails. We're proud to be a day-0 inference provider for NVIDIA NemoClaw, bringing you the speed and efficiency you need. Deploy your agents on Fireworks today! https://lnkd.in/gFqHBPb2