Tele-AI / TeleChat2
星辰语义大模型TeleChat2是由中国电信人工智能研究院研发训练的大语言模型,是首个完全国产算力训练并开源的千亿参数模型
☆123Updated this week
Related projects ⓘ
Alternatives and complementary repositories for TeleChat2
- ☆93Updated 3 months ago
- 😜 表情包视觉数据集,使用glm-4v、step-1v的图像解析能力标注。☆98Updated 6 months ago
- 顾名思义:手搓的RAG☆110Updated 8 months ago
- LLM全栈优质资源汇总☆357Updated 2 months ago
- ☆217Updated last year
- 利用免费的大模型api来结合你的私域数据来生成sft训练数据(妥妥白嫖)支持llamafactory等工具的训练数据格式☆19Updated this week
- hf-mirror-cli 使用国内镜像,无需配置开箱即用,快速下载hugingface上的模型☆119Updated 8 months ago
- ☆49Updated 8 months ago
- TinyRAG☆233Updated 2 weeks ago
- 基于ReAct手搓一个Agent Demo☆105Updated 6 months ago
- AM (Advanced Mathematics) Chat is a large language model that integrates advanced mathematical knowledge, exercises in higher mathematics…☆164Updated 3 months ago
- HF🤗每日简报机器人☆34Updated this week
- FlagEval is an evaluation toolkit for AI large foundation models.☆299Updated 4 months ago
- 训练一个对中文支持更好的LLaVA模型,并开源训练代码和数据。☆38Updated 2 months ago
- ☆213Updated 6 months ago
- Alpaca Chinese Dataset -- 中文指令微调数据集【人工+GPT4o持续更新】☆184Updated last month
- 中文Mixtral-8x7B(Chinese-Mixtral-8x7B)☆641Updated 2 months ago
- ☆78Updated 11 months ago
- Phi2-Chinese-0.2B 从0开始训练自己的Phi2中文小模型,支持接入langchain加载本地知识库做检索增强生成RAG。Training your own Phi2 small chat model from scratch.☆483Updated 4 months ago
- ☆93Updated 5 months ago
- 更纯粹、更高压缩率的Tokenizer☆449Updated 6 months ago
- 从0到1构建一个MiniLLM (pretrain+sft+dpo实践中)☆322Updated 2 months ago
- 「大模型」3小时从0训练27M参数的视 觉多模态VLM,个人显卡即可推理训练!☆341Updated last week
- 面向中文大模型价值观的评估与对齐研究☆476Updated last year
- Skywork-MoE: A Deep Dive into Training Techniques for Mixture-of-Experts Language Models☆127Updated 5 months ago
- LingoWhale-8B: Open Bilingual LLMs | 开源双语预训练大模型☆133Updated last year
- 从0开始,将chatgpt的技术路线跑一遍。☆145Updated 2 months ago
- A streamlined and customizable framework for efficient large model evaluation and performance benchmarking☆236Updated this week
- 欢迎来到 "LLM-travel" 仓库!探索大语言模型(LLM)的奥秘 🚀。致力于深入理解、探讨以及实现与大模型相关的各种技术、原理和应用。☆266Updated 3 months ago
- Llama3-Chinese是以Meta-Llama-3-8B为底座,使用 DORA + LORA+ 的训练方法,在50w高质量中文多轮SFT数据 + 10w英文多轮SFT数据 + 2000单轮自我认知数据训练而来的大模型。☆289Updated 6 months ago