Fine-tuning
Train and host custom models on your own data.
33 tools
Fine-tuning has gone from "deep ML team only" to "a few hours of JSONL away" — but the choice between closed-model FT (OpenAI), open-model FT (Together, Modal), and memory-tuning matters more than ever.
Covers closed-model fine-tuning (OpenAI), open-model FT + serving (Together AI, Replicate, Modal), distributed training platforms (Anyscale), and specialised platforms (Lamini for factual recall).
Pick OpenAI for the easiest UX on closed models. Pick Together AI for open-model FT + serving in one place. Pick Modal for serverless GPU control. Pick Lamini specifically for hallucination-free factual recall.
Ray Tune
Open-source Python library for distributed hyperparameter tuning at any scale.
RunPod
On-demand GPU cloud and serverless inference platform built specifically for AI workloads.
SGLang
Open-source high-throughput inference engine for LLMs and multimodal models with OpenAI-compatible serving.
Scale GenAI Platform
Enterprise agent platform from Scale AI that connects your data, orchestrates multi-agent workflows, and learns from human feedback inside your own VPC.
Together AI Fine-tuning
Managed fine-tuning platform for open-source LLMs and vision models with LoRA, full fine-tuning, and RL support.
Unsloth
Open-source LLM fine-tuning toolkit with custom kernels that train 2-30x faster and use up to 90% less VRAM.
Velda
Serverless GPU orchestration that runs AI training and batch jobs without Docker or Kubernetes.
W&B Sweeps
Hyperparameter optimization from Weights & Biases with Bayesian search and Hyperband early stopping.
vLLM
Open-source high-throughput inference engine for serving LLMs with PagedAttention and continuous batching.