W&B Serverless SFT
Serverless supervised fine-tuning for LLMs on CoreWeave's managed GPU cluster, run alongside Serverless RL in a unified workflow via the Agent Reinforcement Trainer (ART) API.
Find alternatives to W&B Serverless SFT
No pricing published on the product's own page or on wandb.ai/site/pricing; the page's only CTA links to documentation, not a rate card.
What it does
- Model training
- GPU cloud
Sources
- Pricing
-
Serverless SFT is listed under the Platform menu on the pricing page but carries no rate card there.
- Description
-
Official documentation · 27 Sep 2026
"Serverless SFT provides on-demand GPU capacity on the CoreWeave cloud as a serverless backend for the Agent Reinforcement Trainer (ART) API." Third sibling under the Serverless Training umbrella alongside Serverless RL (post-training) and Serverless Inference (serving): "switch quickly between SFT and RL without shuttling artifacts across systems."
- What it does
- Sold within
-
No standalone entitlement or add-on statement found; described only as 'a serverless backend for the Agent Reinforcement Trainer (ART) API', same unpackaged shape as sibling records wandb-serverless-rl and wandb-serverless-inference.
- Name
- URL
Also from Weights & Biases
-
W&B Models Platform
Experiment tracking for model training runs, recording hyperparameters, metrics and artifacts so runs can be compared, swept and reproduced.
-
W&B Weave Developer tool
Tracing, evaluation and production monitoring for LLM and agent applications, capturing each call so prompts and outputs can be scored over time.
-
W&B Registry Platform
Curated central repository providing versioning, aliases, lineage tracking and governance for models and datasets across the ML lifecycle.
-
W&B Serverless Inference API service
Hosted inference service for open-source and commercial LLMs (OpenAI, Qwen, Llama, Kimi, Phi, DeepSeek, Z.AI) without managing infrastructure.
-
W&B Serverless RL Platform
Managed reinforcement-learning fine-tuning service for LLMs on CoreWeave's managed GPU cluster, billed per-token for rollouts with automatic scale-to-zero.
Open Weights & Biases in the directory
Something wrong here? Send a correction — quote this product id: wandb-serverless-sft.