Post
32
π REV LLM: Open Models, Reworked with Intent
REV LLM is an independent project focused on improving open-weight language and multimodal models through fine-tuning, LoRA training, distillation, and model merging.
Rev-Qwen3.5-2B
Rev-Qwen3.5-2B is the first release, based on Qwen/Qwen3.5-2B.
The model focuses on communication, translation, mathematics, science, medicine, coding, and agent functions.
- Model: ehristoforu/rev-qwen3.5-2b
- Collection: https://huggingface.co/collections/ehristoforu/rev-llm-models
- License: Apache-2.0, inherited from the base model
The release combines fine-tuned model merging with a LoRA trained through distillation from open-weight flagship models.
REV LLM is an independent project focused on improving open-weight language and multimodal models through fine-tuning, LoRA training, distillation, and model merging.
Rev-Qwen3.5-2B
Rev-Qwen3.5-2B is the first release, based on Qwen/Qwen3.5-2B.
The model focuses on communication, translation, mathematics, science, medicine, coding, and agent functions.
- Model: ehristoforu/rev-qwen3.5-2b
- Collection: https://huggingface.co/collections/ehristoforu/rev-llm-models
- License: Apache-2.0, inherited from the base model
The release combines fine-tuned model merging with a LoRA trained through distillation from open-weight flagship models.