Fireworks announced general availability of its Training API and Fireworks Lab, aiming to make model specialization accessible beyond labs that already own training clusters. The API lets teams connect a custom Python training loop to Fireworks-managed distributed training and rollout infrastructure, keeping control of loss, reward, data, and environment. Compute is split into Serverless mode for fast experiments and Dedicated mode when scale, control, and GPU economics matter. The stack is built for reinforcement learning as a single train-sample-serve-retrain loop, with three hard requirements: rollout throughput, weight sync so fresh weights land on the rollout deployment every step, and correctness so trainer and inference engine agree on the same model. Fireworks Lab additionally embeds forward-deployed researchers and engineers to co-design or build a model. Named customers include Cursor, Harvey, Figma, Vercel, Cognition, Factory, and Heidi.
Key Takeaways
- βThe Training API attaches a custom Python loop to managed distributed training and rollout, with Serverless and Dedicated compute.
- βThe RL loop is optimized for rollout throughput, per-step weight sync, and trainer-inference model agreement.
- βFireworks Lab embeds researchers with customers including Cursor, Harvey, Figma, Vercel, Cognition, and Factory.