Skip to main content
You can 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.

​
Experimental models

The following models are available as experimental training targets. They may not appear in your model selector by default, and support can change without notice. They are available on request.
  • Qwen/Qwen3.5-4B
  • Qwen/Qwen3.5-9B
  • Qwen/Qwen3.6-35B-A3B
  • Qwen/Qwen3.8-27B
  • google/gemma-4-26B-A4B-it
  • google/gemma-4-31B-it
These lists apply 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 in advance

Train a few promising models, evaluate them against the behavior you want, and deploy the strongest result. This lets you compare results for your task instead of choosing a model based only on benchmarks or size.
Copy a listed value exactly into base_model when creating a fine-tune:
Use the complete request in Fine-tuning Quick Start.
Last modified on September 2, 2026