Gravity-30B-A5B-Preview is a post-trained language model built on Gravity-30B-A5B-Base by Trillion Labs. Starting from the 128K-context base model, it underwent long-context supervised fine-tuning (SFT) and several rounds of preference optimization (DPO), with the final rounds targeting multi-turn tool use using preference labels computed by executing the model’s tool calls rather than by an LLM judge. The model is a hybrid thinking model: it reasons inside … by default and can be switched to direct answers per request. This is a preview release. The checkpoint uses the GravityMoE architecture (trustremotecode=True), whose modeling code inherits directly from the DeepSeek-V3 implementation in transformers. For full architectural details (MLA, MoE routing, tokenizer, etc.), see the base model card. 1. Long-Context SFT — Instruction tuning at the full 128K context (sequence packing) on a mixture of general chat, reasoning, science, code and agentic trajectories. 2. Preference Optimization (DPO) — A round of DPO on judged preference pairs targeting response quality and termination behaviour. 3. Execution-Labelled Multi-Turn DPO — Two rounds of DPO for multi-turn tool use. Candidate…
Модальности:
Генерация текста
Области применения:
Диалог / чат Вызов функций (Tool use)
Задача: Генерация текста
Автор: trillionlabs
Теги: gravity_moe, moe, mixture-of-experts, gravity, trillion-labs, chat, post-trained, function-calling
Лайков: 3 | Загрузок: 531
Описание основано на материалах HuggingFace. Перевод выполнен автоматически.
