Skip to main contentLLMTune publishes upcoming functionality so teams can plan ahead. Timelines may shift, but the sequence reflects the current product strategy.
In Development
- Image-to-Text (Vision) – Upload screenshots, UI flows, and visual tickets to fine-tune multimodal assistants.
- Embeddings + Retrieval – Bring your own knowledge base and query it through retrieval-augmented generation pipelines.
- Instruction Templates – Guided workflows for tone alignment, support copilots, and structured response formats.
- Improved Usage Dashboards – Deeper analytics for per-endpoint latency, error breakdowns, and budget alerts.
Planned
- Code-Aware Fine-Tuning – Models specialized for code generation, code review, and repository Q&A.
- Team-level Permissions – Granular roles for dataset annotators versus deployment operators.
- Secrets Management – Store third-party API keys used by custom inference workflows.
- Self-hosted Agents – Export fine-tuned models with runtime configuration for on-prem orchestration.
Recently Shipped
- Fine-tuning studio with guided three-step flow.
- Managed inference endpoints with role-based API keys.
- Usage metering with requests, tokens, and spend dashboards.
- Webhook events for training and deployment lifecycles.
To request features or share feedback, open a ticket from the dashboard or contact the product team. 
