Unverified90% confidenceFactTime unknown
LLaMA-Factory和Unsloth是社区中成熟的LoRA微调和SFT工具
1
Sources
90%
Confidence
Long-term
Relevance
6/1/2026
First Seen
Sources
Related Entities
Related Claims
UnverifiedWith parameter-efficient fine-tuning methods such as LoRA and QLoRA, as well as no-code platforms such as Coze and ModelScope community, models can be directionally optimized for specific scenarios even without GPU resources or a deep learning background.61% similarUnverifiedA complete technology stack has formed around models such as DeepSeek, including inference frameworks (vLLM, SGLang) and Agent frameworks (such as domestic adaptations of MetaGPT and AutoGen)57% similarVerifiedUnsloth supports fine-tuning mainstream open-source large models including Gemma 4, Qwen3, DeepSeek, and gpt-oss.57% similarUnverified开源社区对o1的复现尝试包括DeepSeek-R1和QwQ，通过强化学习训练模型的推理过程是实现类似能力的关键路径57% similarUnverifiedDeepSeek's DFlash technology sparked community interest in MTP support.55% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/35397API
curl https://kongchang.com/api/v1/knowledge/claims/35397MCP
get_claim(id=35397)