Fine-Tuning - Forii — India's Sovereign Inference Platform

This endpoint is not yet available. It is planned for a future release.

Fine-tune models on Hindi, Tamil, Bengali, and other Indic-language data. Matches OpenAI’s fine-tuning API shape so existing training pipelines work.

Planned endpoints

Operation	Method	Path
Create SFT job	POST	`/inference/v1/fine_tuning/jobs`
List SFT jobs	GET	`/inference/v1/fine_tuning/jobs`
Get SFT job	GET	`/inference/v1/fine_tuning/jobs/{id}`
Cancel SFT job	POST	`/inference/v1/fine_tuning/jobs/{id}/cancel`

No global platform offers fine-tuning on Indic-language data well. Forii will provide:

Hindi, Tamil, Bengali — Domain-specific fine-tuning on Indic languages
Legal, medical, financial — Vertical-specific datasets for Indian industries
SFT first, DPO later — Start with supervised fine-tuning, add DPO and RFT later

DPO (Direct Preference Optimization) and RFT (Reinforcement Fine-Tuning) are planned for a future release, after SFT is solid.