🚀 GLM-4-32B-0414

The GLM-4-32B-0414 series models offer high performance comparable to leading models and support user - friendly local deployment.

⚠️ Important Note

<strong><a href="https://docs.unsloth.ai/basics/unsloth-dynamic-v2.0-gguf">Unsloth Dynamic 2.0</a> achieves superior accuracy & outperforms other leading quants.</strong>

🚀 Quick Start

This section is not provided in the original README, so it is skipped.

✨ Features

Model Introduction

The GLM family welcomes new members, the GLM-4-32B-0414 series models, featuring 32 billion parameters. Its performance is comparable to OpenAI’s GPT series and DeepSeek’s V3/R1 series. It also supports very user-friendly local deployment features. GLM-4-32B-Base-0414 was pre-trained on 15T of high-quality data, including substantial reasoning-type synthetic data. This lays the foundation for subsequent reinforcement learning extensions. In the post-training stage, human preference alignment was employed for dialogue scenarios. Additionally, using techniques like rejection sampling and reinforcement learning, the model’s performance in instruction following, engineering code, and function calling was enhanced, thus strengthening the atomic capabilities required for agent tasks. GLM-4-32B-0414 achieves good results in engineering code, Artifact generation, function calling, search-based Q&A, and report generation. In particular, on several benchmarks, such as code generation or specific Q&A tasks, GLM-4-32B-Base-0414 achieves comparable performance with those larger models like GPT-4o and DeepSeek-V3-0324 (671B).

GLM-Z1-32B-0414 is a reasoning model with deep thinking capabilities. This was developed based on GLM-4-32B-0414 through cold start, extended reinforcement learning, and further training on tasks including mathematics, code, and logic. Compared to the base model, GLM-Z1-32B-0414 significantly improves mathematical abilities and the capability to solve complex tasks. During training, general reinforcement learning based on pairwise ranking feedback was also introduced, which enhances the model's general capabilities.

GLM-Z1-Rumination-32B-0414 is a deep reasoning model with rumination capabilities (against OpenAI's Deep Research). Unlike typical deep thinking models, the rumination model is capable of deeper and longer thinking to solve more open-ended and complex problems (e.g., writing a comparative analysis of AI development in two cities and their future development plans). Z1-Rumination is trained through scaling end-to-end reinforcement learning with responses graded by the ground truth answers or rubrics and can make use of search tools during its deep thinking process to handle complex tasks. The model shows significant improvements in research-style writing and complex tasks.

Finally, GLM-Z1-9B-0414 is a surprise. All the aforementioned techniques were employed to train a small model (9B). GLM-Z1-9B-0414 exhibits excellent capabilities in mathematical reasoning and general tasks. Its overall performance is top-ranked among all open-source models of the same size. Especially in resource-constrained scenarios, this model achieves an excellent balance between efficiency and effectiveness, providing a powerful option for users seeking lightweight deployment.

Showcase

Animation Generation

Web Design

Model	Demo	Prompt
GLM-4-32B-0414		Design a drawing board that supports custom function plotting, allowing adding and deleting custom functions, and assigning colors to functions.
GLM-4-32B-0414		Design a UI for a mobile machine learning platform, which should include interfaces for training tasks, storage management, and personal statistics. The personal statistics interface should use charts to display the user's resource usage over a period. Use Tailwind CSS to style the page, and display these 3 mobile interfaces tiled on a single HTML page.

SVG Generation

Model	Demo	Prompt
GLM-4-32B-0414		Create a misty Jiangnan scene using SVG.
GLM-4-32B-0414		Use SVG to illustrate the training process of an LLM.

Search-Based Writing

For search-based writing tasks, the following system prompt is used to have the model respond based on search results:

Please answer the user's question based on the given search results.

## Note
1. Make full use of and organize the collected information, rather than simply copying and pasting, to generate in-depth and professional answers that meet the user's requirements.
2. When sufficient information is provided, your answer should be as long as possible. From the perspective of the user's intention, provide a response with sufficient information and multiple perspectives.
3. Additionally, not all search results are closely related to the user's question. Please carefully甄别, screen, and utilize them.
4. The answers to objective questions are usually very short. You can appropriately add one or two related sentences to enrich the content.
5. Please ensure that your response is beautifully formatted and highly readable. For multi-entity comparisons or listings, make good use of the list format to help the user better understand the information.
6. Unless the user requests otherwise, please keep the language of your answer consistent with the language of the user's question.
7. When appropriate, use the format such as:【0†source】at the end of the sentence to cite the search results.

When using, you can obtain search results through methods such as RAG or WebSearch, and wrap them in observation, for example:

[
    {
        "role": "user",
        "content": "Explore the common characteristics of children's literature, with a focus on its narrative techniques and thematic tendencies. This includes narrative techniques: common approaches in children's literature such as first-person, third-person, omniscient narrator, and interactive narration, and their influence on young readers. It also includes thematic tendencies: recurring themes in children's literature such as growth, adventure, friendship, and family, with an analysis of how these themes impact children's cognitive and emotional development. Additionally, other universal features such as the use of personification, repetitive language, symbolism and metaphor, and educational value should be considered. Please provide a detailed analytical report based on academic research, classic examples of children's literature, and expert opinions."
    },
    {
        "role": "observation",
        "content": "【{id}†{title}†{url}】\n{content}"
    },
    ...
]

For the above prompt, an internal or external search model is used to obtain the search results. Using the format shown above, the following analysis report can be generated:

📄 License

The project is released under the MIT license.