ChatGLM: A Family of Large Language Models from GLM-130B to GLM-4 All Tools
By Team Glm Aohan Zeng, Bin Xu, Bowen Wang, Chenhui Zhang, Da Yin, Diego Rojas, Guanyu Feng, Hanlin Zhao, Hanyu Lai, Hao Yu and 44 more. 2024
1,646 citations recorded by the source corpus.
We introduce ChatGLM, an evolving family of large language models that we have been developing over time. This report primarily focuses on the GLM-4 language series, which includes GLM-4, GLM-4-Air, and GLM-4-9B. They represent our most capable models that are trained with all the insights and lessons gained from the preceding three generations of ChatGLM. To date, the GLM-4 models are pre-trained on ten trillions of tokens mostly in Chinese and English, along with a small set of corpus from 24 languages, and aligned primarily for Chinese and English usage. The high-quality alignment is achieved via a multi-stage post-training process, which involves supervised fine-tuning and learning from human feedback. Evaluations show that GLM-4 1) closely rivals or outperforms GPT-4 in terms of general metrics such as MMLU, GSM8K, MATH, BBH, GPQA, and HumanEval, 2) gets close to GPT-4-Turbo in instruction following as measured by IFEval, 3) matches GPT-4 Turbo (128K) and Claude 3 for long context tasks, and 4) outperforms GPT-4 in Chinese alignments as measured by AlignBench. The GLM-4 All Tools model is further aligned to understand user intent and autonomously decide when and which tool(s) touse -- including web browser, Python interpreter, text-to-image model, and user-defined functions -- to effectively complete complex tasks. In practical applications, it matches and even surpasses GPT-4 All Tools in tasks like accessing online information via web browsing and solving math problems using Python interpreter. Over the course, we have open-sourced a series of models, including ChatGLM-6B (three generations), GLM-4-9B (128K, 1M), GLM-4V-9B, WebGLM, and CodeGeeX, attracting over 10 million downloads on Hugging face in the year 2023 alone. The open models can be accessed through https://github.com/THUDM and https://huggingface.co/THUDM.
Subfields: Artificial Intelligence, Large Language Models, LLM Agents & Tool Use
How researchers see this paper
The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.
In Artificial Intelligence
In Large Language Models
Citations in the corpus
- RULER: What's the Real Context Size of Your Long-Context Language Models? (2024, background): “…2024a; Bulatov et al., 2023; Martins et al., 2022; Wu et al., 2022), retrieving relevant information from context (Xu et al., 2024a; Mohtashami & Jaggi, 2023; Wang et al., 2024; Tworkowski et al., 2024; Xiao et al., 2024a), or preserving the salient information via compression (Jiang et al., 2023).”
- LlamaFactory: Unified Efficient Fine-Tuning of 100+ Language Models (2024, neutral): “LLaMA-Adapter (Zhang et al., 2024) efficiently fine-tunes the Llama model (Touvron et al., 2023a) using a zero-initialized attention.”
- Qwen2 Technical Report (2024, background): “Qwen2-7B-Instruct achieves parity with ChatGLM4-9B-1M, albeit with a more noticeable decline at extended contexts.”
- SWIFT:A Scalable lightWeight Infrastructure for Fine-Tuning (2024, neutral): “5-VL [2], GLM4-V [19], InternVL3 [65], and DiT models [45] , can all satisfy to this trend.”
- GLM-4.5V and GLM-4.1V-Thinking: Towards Versatile Multimodal Reasoning with Scalable Reinforcement Learning (2025, neutral): “For the LLM component, we use GLM-4-9B-0414 [13] for the GLM-4.”
- MInference 1.0: Accelerating Pre-filling for Long-Context LLMs via Dynamic Sparse Attention (2024, positive): “…release update of Qwen series model with up to 128K context window that achieve superior or comparable performance compared to LLaMA-3; 6) GLM-4-9B-1M [GZX + 24] has been improved from its predecessor in terms of a 1M context window, performance on downstream tasks and inference efficiency.”
- HunyuanVideo: A Systematic Framework For Large Video Generative Models (2024, neutral): “We have configured HunyuanVideo with a series of MLLMs [78, 17, 26] for different purposes.”
- WizardMath: Empowering Mathematical Reasoning for Large Language Models via Reinforced Evol-Instruct (2023, neutral): “…with models like OpenAI’s GPT Series (Brown et al., 2020a; OpenAI, 2023), Anthropic’s Claude (Bai et al., 2022), Google’s PaLM (Chowdhery et al., 2022; Anil et al., 2023), Gemini (Team, 2023), and Gemma (Team et al., 2024) featuring billions of parameters and trained on massive textual datasets.”
- RUMAD: Reinforcement-Unifying Multi-Agent Debate (2026, unknown): “1-8B-Instruct [21], ChatGLM-4-9B [5], and Deepseek-Math-7B-Instruct [6].”
- CFVBench: A Comprehensive Video Benchmark for Fine-grained Multimodal Retrieval-Augmented Generation (2025, unknown): “Majority voting was applied, and Fleiss’s kappa [14]”
- ExpandR: Teaching Dense Retrievers Beyond Queries with LLM Guidance (2025, unknown): “Recent advances in LLMs (Achiam et al., 2023; GLM et al., 2024) offer new opportunities to address this issue by leveraging their rich knowledge and powerful generative capabilities to enrich or reformulate user queries (Yu et al., 2020; Lin et al., 2020; Ye et al., 2023).”
- LLM-confidence reranker: A training-free approach for enhancing retrieval-augmented generation systems (2026, unknown): “1-8B-Instruct (Llama8B) (Dubey et al., 2024), GLM-4-9B-Chat (GLMChat9B) (Zeng et al., 2024), and InternLM2.”
- MAGECODE: Machine-Generated Code Detection Method Using Large Language Models (2024, unknown): “…have been demonstrated in prior studies [3] to be rather effective in identifying machine-generated content from six LLMs, including ChatGLM [16], Dolly [17], ChatGPT-turbo [7], 190188 VOLUME 12, 2024 Authorized licensed use limited to the terms of the applicable license agreement with IEEE.”