SupraLabs releases small model for cheaper chat title generation
SupraLabs released `supra-title-FFT-preview`, a single-purpose model that turns a chat message into a short title. Its earlier title model was trained on 12,000 examples, worked reasonably well on common chats, and struggled more with niche topics. The new release uses a cleaner filtered dataset with 115,000 chat-title examples, giving it broader coverage.
The base model is `LiquidAI/LFM2.5-350M-Base`, with about 0.4 billion . The model was trained with ``, not `LoRA`. It is designed for one job only: send the user message and get a title back, with no needed.
The related dataset can also be used for instruction tuning, title-generation testing, or -style title experiments with small models.
Key points
- `supra-title-FFT-preview` is built only for chat title generation.
- The set grew from 12,000 examples to 115,000 filtered examples.
- The model is small, using about 0.4 billion .
- It can generate a title from the user message without a .
- It may reduce for routine agent-side tasks like conversation naming.