ModelsProviders
OpenAI
Fine-tune GPT-4.1 and GPT-4.1 Mini on Commissioned.
OpenAI models are strong all-rounders with excellent instruction following. Fine-tuning runs on OpenAI's cloud infrastructure and takes 30–45 minutes.
Available models
OpenAI's most capable fine-tunable model. Best for complex tasks that require nuanced understanding.
Strengths:
- Highest reasoning quality among OpenAI models
- Excellent at code generation, analysis, and creative writing
- Strong instruction following
Good for: writing assistants, code assistants, complex analysis, multi-step reasoning
Plan: Free
A smaller, faster version of GPT-4.1. The best default choice for most fine-tunes.
Strengths:
- Very fast inference
- Lower cost per token when used via API
- Still very capable for most tasks
Good for: support agents, Q&A bots, document summarization, general-purpose assistants
Plan: Free
Training details
| GPT-4.1 | GPT-4.1 Mini | |
|---|---|---|
| Training time | 30–45 min | 30–45 min |
| Output | Hosted by OpenAI | Hosted by OpenAI |
| Self-hosting | Not available | Not available |
| Plan required | Free | Free |