Models you can train - CoreWeave Docs

You can currently train:

OpenPipe/Qwen3-14B-Instruct
Qwen/Qwen3.6-27B
Qwen/Qwen3-30B-A3B-Instruct-2507
meta-llama/Llama-3.1-8B-Instruct

The best model depends on your task. A smaller model may be faster and cheaper, while a larger model may handle more difficult behavior. Model Distillation makes it practical to train several candidates and compare them on the same examples.

This list applies to models you can train. You can also connect existing models from OpenAI-compatible providers for serving, relabeling, and comparison.

You do not need to choose one model upfront

Train a few promising models, evaluate them against the behavior you want, and deploy the strongest result. This is often more reliable than choosing a model based only on benchmarks or size. Automation can run this search for you by training multiple candidates and promoting only a model that meets your quality goal.

API: Use model IDs

Copy a listed value exactly into base_model when creating a fine-tune, or add one model object per candidate in Automation:

{
  "models": [
    {"base_model": "Qwen/Qwen3.6-27B"},
    {"base_model": "meta-llama/Llama-3.1-8B-Instruct"}
  ]
}

Use the complete requests in Fine-tuning Quick Start or Automation.

Last modified on August 25, 2026

​You do not need to choose one model upfront

You do not need to choose one model upfront