No GRPO or reinforcement fine-tuning for reasoning models
Users cannot run GRPO or other reinforcement-learning fine-tuning to train reasoning models like DeepSeek-R1 in the tool.
Tier II · Emerging Problems that are still building up.
3 public reports · first reported 24 October 2024 · last reported 2 February 2025
What's going wrong
Users who want to reproduce reasoning models in the style of DeepSeek-R1 report that the fine-tuning tool has no reinforcement learning path such as GRPO.
AI-assisted: written with the help of AI from the reports. The idea report has the full summary.
In the idea report
- The full summary.
- The products the 3 reports are about, and how many came each month.
- Related problems.
No business has been found with this problem yet, so the report has no contacts. All sales are final — no refunds. Pricing.