Views
No views yet
DeepSeek-R1-0528 + (RakutenAI-3.0 - DeepSeek-V3-0324)DeepSeek-R1-0528 with the Japanese language expertise of RakutenAI-3.0, while subtracting the base DeepSeek-V3-0324 to isolate task-specific improvements.| Parameter | Value |
|---|---|
| Base Model | DeepSeek-R1-0528 |
| Task Vector Source | RakutenAI-3.0 - DeepSeek-V3-0324 |
| Architecture | Mixture of Experts (MoE) |
| Context Length | 128K tokens |
| License | Apache-2.0 |
1from transformers import AutoModelForCausalLM, AutoTokenizer
2
3model = AutoModelForCausalLM.from_pretrained("Local-Novel-LLM-project/RAI-3.0-R1-VECTOR", trust_remote_code=True)
4tokenizer = AutoTokenizer.from_pretrained("Local-Novel-LLM-project/RAI-3.0-R1-VECTOR")
5
6inputs = tokenizer("日本の文化で重要な要素は", return_tensors="pt")
7outputs = model.generate(**inputs, max_length=100)
8print(tokenizer.decode(outputs[0]))1@misc{RAIR1VECTOR2026,
2 title = {RAI-3.0-R1-VECTOR: Task-Vector Merged Model},
3 author = {LocalNovelLLM-project},
4 year = {2026},
5 publisher = {LocalNovelLLM-project},
6 url = {https://huggingface.co/Local-Novel-LLM-project/RAI-3.0-R1-VECTOR}
7}