chatglm2 6b

We have hosted the application chatglm2 6b in order to run this application in our online workstations with Wine or directly.

Quick description about chatglm2 6b:

ChatGLM2-6B is the second-gen Chinese-English conversational LLM from ZhipuAI/Tsinghua. It upgrades the base model with GLM’s hybrid pretraining objective, 1.4 TB bilingual data, and preference alignment—delivering big gains on MMLU, CEval, GSM8K, and BBH. The context window extends up to 32K (FlashAttention), and Multi-Query Attention improves speed and memory use. The repo includes Python APIs, CLI & web demos, OpenAI-style/FASTAPI servers, and quantized checkpoints for lightweight local deployment on GPUs or CPU/MPS.

Features:

• Stronger base model: large bilingual pretrain + alignment; big benchmark lifts
• Long context variants: 8K default, 32K model (LongBench-competitive)
• Faster, lighter inference: Multi-Query Attention + FlashAttention
• Low-cost deploy: FP16/BF16, INT8/INT4 (?5.5 GB), CPU & Apple MPS
• Demos & APIs: CLI, Gradio/Streamlit, FastAPI and OpenAI-format servers
• Finetuning & tooling: P-Tuning v2, full-parameter scripts, multi-GPU utilities

Programming Language: Python, Unix Shell.
Categories: