Together AI Fine-Tuning Platform (Expert LoRA / DPO update)
Together AI's hosted fine-tuning platform adds a no-code UI, Direct Preference Optimization, continued training, and Expert LoRA for MoE models.
Overview
On September 11, 2026, Together AI shipped a major update to its hosted fine-tuning platform, adding a no-code web UI (previously SDK/API-only), Direct Preference Optimization and continued-training support, and 'Expert LoRA' -- a Mixture-of-Experts-specific low-rank adaptation technique for fine-tuning sparse MoE models like GLM-5.3. New model support includes Gemma 3 and distilled DeepSeek-R1 variants.
This is a sourced directory entry. Features and capabilities are reported in the linked sources; this entry does not document independent Aymorphic testing.
Key features
- No-code browser fine-tuning UI (previously code/API only)
- Direct Preference Optimization support
- Continued training on top of prior fine-tunes
- Expert LoRA for MoE models (e.g. GLM-5.3)
- New model support: Gemma 3, distilled DeepSeek-R1 variants
Use cases
- Fine-tuning open-weight models without owning GPU infrastructure
- MoE-specific adaptation
- Preference-tuning for alignment
Pricing
unknown (usage-based per Together AI's standard model, not itemized in sources).
Pricing reflects the source record and may change. Confirm current terms with the provider.
Dates and sources
- Released
- 11 Sept 2026
- Record updated
- 23 Sept 2026
- Record last checked
- 23 Sept 2026