
The next frontier of competitive advantage is specialized intelligence. The most ambitious companies are training models with the capabilities that underpin differentiated products and operations, across coding, chip design, cybersecurity, and beyond:
Across these workloads, training a base model for specific tasks improved quality, latency, cost, and/or behavioral consistency. Fireworks makes this level of specialization accessible to teams through three training surfaces, each designed for a different level of expertise, operational ownership, and control.
| Training API | Managed Training | Fireworks Lab | |
|---|---|---|---|
| Built for | ML Researchers | ML Engineers | All Levels |
| Workflow | Bring your own data and loop for deepest control. From LoRA and per-token pricing to full-parameter and reserved capacity. | Bring your data or evals, we run the loop. Pick your method (SFT, RFT, DPO) and run via UI/API. | We provide embedded ML researchers and forward-deployed engineers. Engagements range from co-design to custom build. |

No infrastructure to manage: Choose your method on Managed Training or write your own loop with the Training API.
Today, we are announcing the general availability of our Training API and Fireworks Lab.
We’ve worked with ML teams around the world to understand where existing training workflows fall short: constrained model and method choice, limited control over parameters and training loops, rigid and underutilized compute, and fragmented training and rollout infrastructure that bloats costs, creates synchronization overhead, and introduces numerical drift. Those partnerships shaped our Training API.
The Fireworks Training API connects your custom training loop to Fireworks-managed distributed training and rollout infrastructure. You orchestrate the loop in Python from wherever you choose, with full control over your loss or reward, data, and environment. Fireworks manages high-performance compute: the trainer computing gradients and updating the model, and the rollout deployment generating samples from the trained model. Fireworks also manages the interaction between these two, from weight synchronization to failed-swap recovery, and train-rollout alignment. All of this allows you to focus on refining your learning signal, not infrastructure challenges.
pythonCopy1234567
Example of a custom reward function defined in Python. This code can be dropped directly into a training loop using the Fireworks Training API.
Choose Serverless when speed to experiment matters most, or Dedicated when scale, control, and GPU economics take priority.
Serverless training allows you to train LoRA adapters on top of shared infrastructure. You pay per token, and sampling runs in the same session. Use it to iterate on experiments, de-risk a larger run, or run RL loops that alternate between rollout and training.

Deploy a promising checkpoint to production inference instantly through the UI (Managed Training) or API (Training API & Managed Training). With multi-LoRA deployment, each customer or use case gets its own tuned model without standing up separate infrastructure.
Dedicated training is best when you need full-parameter training, models beyond the serverless pool, larger context lengths or LoRA ranks, or sustained throughput where per-GPU-hour pricing is more economical than per-token. Fireworks gives you dedicated, elastic capacity sized to each run, with the ability to scale GPU type and cluster size as needed.
| Serverless | Dedicated | |
|---|---|---|
| Parameter mode | LoRA | LoRA and full-parameter |
| Models | A curated list of popular models | Unlimited depth and breadth, up to the largest MoE models |
| Provisioning | Attach to an always-on shared pool, nothing to spin up | Trainer and deployment provisioned for your run; support for disaggregation to speed up training across trainer or rollout |
| Throughput | Shared capacity and per-account rate limits | No contention, scale up as you need |
| Billing | Per-token | Per-GPU-hour |
Start on Serverless in minutes, nothing to spin up: Start training →
Fireworks is among the few organizations outside the frontier labs to have operated reinforcement learning (RL) across more than 10,000 GPUs. RL turns training and inference into one continuously coupled system; every step starts with rollouts from the current policy, updates the weights, and feeds those weights back into inference for the next round of sampling. At frontier scale, the algorithm is only part of the challenge. Effective RL hinges on three requirements: correctness, performance efficiency, and development speed.
For correctness in RL training, the rollout engine and trainer need to share the same numerical definition. Small numerical drift can corrupt a learning signal, from token clipping to reward collapse, when everything appears to be working. With Fireworks,
Fireworks validates this alignment directly by running the same sequences through the rollout engine and trainer and measuring train-inference KL divergence (KLD). All models launched on Fireworks training are continually validated to ensure KLD is minimized.

At scale, RL is largely a generation problem. Much of the wall-clock time is spent producing rollouts rather than updating weights, so rollout throughput and GPU utilization become primary determinants of training time and cost. Completion lengths also vary widely, which means rigid batching can leave GPUs idle while a small number of long-running trajectories finish.
Fireworks can run asynchronous RL, overlapping rollout collection with training rather than forcing the two sides of the loop to execute sequentially. Rollout GPUs can begin generating the next batch while the trainer updates on the previous one, with bounded weight staleness available when the algorithm permits more overlap. The objective is simple: keep GPUs doing useful work instead of waiting at synchronization barriers.
Once the loop completes a training step, the updated model must be transmitted and loaded into the rollout deployment. Rather than tearing down the rollout deployment and reloading a full model after every step, Fireworks hot-loads new weights into the running deployment. For full parameter checkpoints, novel compression techniques are used to reduce the bandwidth and storage requirements for weight transmission. An XOR Diff is computed between current and previous weights, and zstd compression is used to reduce the size of this payload. This results in up to a 10x reduction in transmission bandwidth, significantly speeding up the training loop.
The bottleneck to production is often not the training job, but the time between iterations. In a fragmented stack, every cycle means moving checkpoints between training and inference systems, redeploying, reconciling differences, running evals, and starting again.
Fireworks turns that process into a continuous development loop: train → deploy → evaluate → retrain on one platform. Checkpoints move directly into serving, while production traces, evals, and feedback feed the next training run. And because we handle GPU provisioning, capacity, and configuration, customers skip the procurement, model enablement, and infrastructure-sizing work that slows iteration and time to market. Teams report 2–4x more iterations on the same training budget, compressing model development from weeks to hours and accelerating time to production.

With correctness, performance efficiency, and iteration speed built into the platform, teams can focus on their training strategy. The strategy includes the decisions that shape model quality: the training loop, data, rewards, evaluators, and harness design. It also includes the systems choices that determine how the workload should run at scale, such as synchronous versus asynchronous RL, acceptable weight staleness, rollout-to-trainer allocation, parallelism, and checkpoint and hot-load strategy.
Fireworks Lab helps teams make those decisions more confidently and execute faster. Forward-deployed researchers and engineers embed directly with your team. A full engagement starts with a diagnosis that defines the capability, baseline, success criteria, and scope, then moves into a time-boxed implementation built backward from your target date. You keep the production-ready model, evaluation harness, data pipelines, training loop, and recipes.