belztjti / vision

huggingface.co
Total runs: 7
24-hour runs: -1
7-day runs: 4
30-day runs: -4
Model's Last Updated: June 07 2024
feature-extraction

Introduction of vision

Model Details of vision

GLM-4-9B-Chat

GLM-4-9B 是智谱 AI 推出的最新一代预训练模型 GLM-4 系列中的开源版本。 在语义、数学、推理、代码和知识等多方面的数据集测评中,GLM-4-9B 及其人类偏好对齐的版本 GLM-4-9B-Chat 均表现出较高的性能。 除了能进行多轮对话,GLM-4-9B-Chat 还具备网页浏览、代码执行、自定义工具调用(Function Call)和长文本推理(支持最大 128K 上下文)等高级功能。 本代模型增加了多语言支持,支持包括日语,韩语,德语在内的 26 种语言。我们还推出了支持 1M 上下文长度(约 200 万中文字符)的模型。

评测结果

我们在一些经典任务上对 GLM-4-9B-Chat 模型进行了评测,并得到了如下的结果:

Model AlignBench-v2 MT-Bench IFEval MMLU C-Eval GSM8K MATH HumanEval NCB
Llama-3-8B-Instruct 5.12 8.00 68.58 68.4 51.3 79.6 30.0 62.2 24.7
ChatGLM3-6B 3.97 5.50 28.1 66.4 69.0 72.3 25.7 58.5 11.3
GLM-4-9B-Chat 6.61 8.35 69.0 72.4 75.6 79.6 50.6 71.8 32.2
长文本

在 1M 的上下文长度下进行 大海捞针实验 ,结果如下:

needle

在 LongBench-Chat 上对长文本能力进行了进一步评测,结果如下:

leaderboard

多语言能力

在六个多语言数据集上对 GLM-4-9B-Chat 和 Llama-3-8B-Instruct 进行了测试,测试结果及数据集对应选取语言如下表

Dataset Llama-3-8B-Instruct GLM-4-9B-Chat Languages
M-MMLU 49.6 56.6 all
FLORES 25.0 28.8 ru, es, de, fr, it, pt, pl, ja, nl, ar, tr, cs, vi, fa, hu, el, ro, sv, uk, fi, ko, da, bg, no
MGSM 54.0 65.3 zh, en, bn, de, es, fr, ja, ru, sw, te, th
XWinograd 61.7 73.1 zh, en, fr, jp, ru, pt
XStoryCloze 84.7 90.7 zh, en, ar, es, eu, hi, id, my, ru, sw, te
XCOPA 73.3 80.1 zh, et, ht, id, it, qu, sw, ta, th, tr, vi
工具调用能力

我们在 Berkeley Function Calling Leaderboard 上进行了测试并得到了以下结果:

Model Overall Acc. AST Summary Exec Summary Relevance
Llama-3-8B-Instruct 58.88 59.25 70.01 45.83
gpt-4-turbo-2024-04-09 81.24 82.14 78.61 88.75
ChatGLM3-6B 57.88 62.18 69.78 5.42
GLM-4-9B-Chat 81.00 80.26 84.40 87.92

本仓库是 GLM-4-9B-Chat 的模型仓库,支持 128K 上下文长度。

运行模型

使用 transformers 后端进行推理:

import torch
from transformers import AutoModelForCausalLM, AutoTokenizer

device = "cuda"

tokenizer = AutoTokenizer.from_pretrained("THUDM/glm-4-9b-chat",trust_remote_code=True)

query = "你好"

inputs = tokenizer.apply_chat_template([{"role": "user", "content": query}],
                                       add_generation_prompt=True,
                                       tokenize=True,
                                       return_tensors="pt",
                                       return_dict=True
                                       )

inputs = inputs.to(device)
model = AutoModelForCausalLM.from_pretrained(
    "THUDM/glm-4-9b-chat",
    torch_dtype=torch.bfloat16,
    low_cpu_mem_usage=True,
    trust_remote_code=True
).to(device).eval()

gen_kwargs = {"max_length": 2500, "do_sample": True, "top_k": 1}
with torch.no_grad():
    outputs = model.generate(**inputs, **gen_kwargs)
    outputs = outputs[:, inputs['input_ids'].shape[1]:]
    print(tokenizer.decode(outputs[0], skip_special_tokens=True))

使用 VLLM后端进行推理:

from transformers import AutoTokenizer
from vllm import LLM, SamplingParams

# GLM-4-9B-Chat-1M
# max_model_len, tp_size = 1048576, 4

# GLM-4-9B-Chat
from transformers import AutoTokenizer
from vllm import LLM, SamplingParams

# 如果遇见 OOM 现象,建议减少max_model_len,或者增加tp_size
max_model_len, tp_size = 131072, 1
model_name = "THUDM/glm-4-9b-chat"
prompt = [{"role": "user", "content": "你好"}]

tokenizer = AutoTokenizer.from_pretrained(model_name, trust_remote_code=True)
llm = LLM(
    model=model_name,
    tensor_parallel_size=tp_size,
    max_model_len=max_model_len,
    trust_remote_code=True,
    enforce_eager=True,
    # GLM-4-9B-Chat-1M 如果遇见 OOM 现象,建议开启下述参数
    # enable_chunked_prefill=True,
    # max_num_batched_tokens=8192
)
stop_token_ids = [151329, 151336, 151338]
sampling_params = SamplingParams(temperature=0.95, max_tokens=1024, stop_token_ids=stop_token_ids)

inputs = tokenizer.apply_chat_template(prompt, tokenize=False, add_generation_prompt=True)
outputs = llm.generate(prompts=inputs, sampling_params=sampling_params)

print(outputs[0].outputs[0].text)
协议

GLM-4 模型的权重的使用则需要遵循 LICENSE 。

Rhe use of the GLM-4 model weights needs to comply with the LICENSE .

引用

如果你觉得我们的工作有帮助的话,请考虑引用下列论文。

@article{zeng2022glm,
  title={Glm-130b: An open bilingual pre-trained model},
  author={Zeng, Aohan and Liu, Xiao and Du, Zhengxiao and Wang, Zihan and Lai, Hanyu and Ding, Ming and Yang, Zhuoyi and Xu, Yifan and Zheng, Wendi and Xia, Xiao and others},
  journal={arXiv preprint arXiv:2210.02414},
  year={2022}
}
@inproceedings{du2022glm,
  title={GLM: General Language Model Pretraining with Autoregressive Blank Infilling},
  author={Du, Zhengxiao and Qian, Yujie and Liu, Xiao and Ding, Ming and Qiu, Jiezhong and Yang, Zhilin and Tang, Jie},
  booktitle={Proceedings of the 60th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)},
  pages={320--335},
  year={2022}
}

Runs of belztjti vision on huggingface.co

7
Total runs
-1
24-hour runs
2
3-day runs
4
7-day runs
-4
30-day runs

More Information About vision huggingface.co Model

More vision license Visit here:

https://choosealicense.com/licenses/glm-4

vision huggingface.co

vision huggingface.co is an AI model on huggingface.co that provides vision's model effect (), which can be used instantly with this belztjti vision model. huggingface.co supports a free trial of the vision model, and also provides paid use of the vision. Support call vision model through api, including Node.js, Python, http.

belztjti vision online free

vision huggingface.co is an online trial and call api platform, which integrates vision's modeling effects, including api services, and provides a free online trial of vision, you can try vision online for free by clicking the link below.

belztjti vision online free url in huggingface.co:

https://huggingface.co/belztjti/vision

vision install

vision is an open source model from GitHub that offers a free installation service, and any user can find vision on GitHub to install. At the same time, huggingface.co provides the effect of vision install, users can directly use vision installed effect in huggingface.co for debugging and trial. It also supports api for free installation.

vision install url in huggingface.co:

https://huggingface.co/belztjti/vision

Url of vision

Provider of vision huggingface.co

belztjti
ORGANIZATIONS

Other API from belztjti

huggingface.co

Total runs: 6.6K
Run Growth: 0
Growth Rate: 0.00%
Updated:October 03 2025
huggingface.co

Total runs: 3.1K
Run Growth: 0
Growth Rate: 0.00%
Updated:December 25 2025
huggingface.co

Total runs: 2.9K
Run Growth: 0
Growth Rate: 0.00%
Updated:December 30 2025
huggingface.co

Total runs: 1.9K
Run Growth: 0
Growth Rate: 0.00%
Updated:December 23 2025
huggingface.co

Total runs: 1.8K
Run Growth: 0
Growth Rate: 0.00%
Updated:February 02 2026
huggingface.co

Total runs: 1.8K
Run Growth: 1.8K
Growth Rate: 100.00%
Updated:November 29 2025
huggingface.co

Total runs: 1.5K
Run Growth: 0
Growth Rate: 0.00%
Updated:November 06 2025
huggingface.co

Total runs: 1.5K
Run Growth: 0
Growth Rate: 0.00%
Updated:January 26 2026
huggingface.co

Total runs: 741
Run Growth: 0
Growth Rate: 0.00%
Updated:February 14 2026
huggingface.co

Total runs: 218
Run Growth: 211
Growth Rate: 96.79%
Updated:August 27 2024
huggingface.co

Total runs: 208
Run Growth: 155
Growth Rate: 74.52%
Updated:June 13 2024
huggingface.co

Total runs: 163
Run Growth: 0
Growth Rate: 0.00%
Updated:August 16 2024
huggingface.co

Total runs: 69
Run Growth: 0
Growth Rate: 0.00%
Updated:September 01 2024
huggingface.co

Total runs: 20
Run Growth: 0
Growth Rate: 0.00%
Updated:July 31 2024
huggingface.co

Total runs: 15
Run Growth: 11
Growth Rate: 73.33%
Updated:June 09 2024
huggingface.co

Total runs: 15
Run Growth: 11
Growth Rate: 73.33%
Updated:June 09 2024
huggingface.co

Total runs: 12
Run Growth: 8
Growth Rate: 66.67%
Updated:June 09 2024
huggingface.co

Total runs: 12
Run Growth: 0
Growth Rate: 0.00%
Updated:August 02 2024
huggingface.co

Total runs: 11
Run Growth: 5
Growth Rate: 45.45%
Updated:June 08 2024
huggingface.co

Total runs: 10
Run Growth: 0
Growth Rate: 0.00%
Updated:September 27 2024
huggingface.co

Total runs: 8
Run Growth: 1
Growth Rate: 12.50%
Updated:June 08 2024
huggingface.co

Total runs: 8
Run Growth: 3
Growth Rate: 37.50%
Updated:July 04 2024
huggingface.co

Total runs: 8
Run Growth: 3
Growth Rate: 37.50%
Updated:July 24 2024
huggingface.co

Total runs: 8
Run Growth: 3
Growth Rate: 37.50%
Updated:June 07 2024
huggingface.co

Total runs: 8
Run Growth: 4
Growth Rate: 50.00%
Updated:June 09 2024
huggingface.co

Total runs: 7
Run Growth: 2
Growth Rate: 28.57%
Updated:June 09 2024
huggingface.co

Total runs: 6
Run Growth: -1
Growth Rate: -16.67%
Updated:June 07 2024
huggingface.co

Total runs: 5
Run Growth: -2
Growth Rate: -40.00%
Updated:June 08 2024
huggingface.co

Total runs: 3
Run Growth: 0
Growth Rate: 0.00%
Updated:January 31 2025
huggingface.co

Total runs: 1
Run Growth: -4
Growth Rate: -400.00%
Updated:June 02 2026
huggingface.co

Total runs: 0
Run Growth: 0
Growth Rate: 0.00%
Updated:July 06 2024
huggingface.co

Total runs: 0
Run Growth: 0
Growth Rate: 0.00%
Updated:June 12 2024
huggingface.co

Total runs: 0
Run Growth: 0
Growth Rate: 0.00%
Updated:February 04 2026
huggingface.co

Total runs: 0
Run Growth: 0
Growth Rate: 0.00%
Updated:January 29 2025
huggingface.co

Total runs: 0
Run Growth: 0
Growth Rate: 0.00%
Updated:June 11 2024
huggingface.co

Total runs: 0
Run Growth: 0
Growth Rate: 0.00%
Updated:July 21 2024
huggingface.co

Total runs: 0
Run Growth: 0
Growth Rate: 0.00%
Updated:December 13 2025
huggingface.co

Total runs: 0
Run Growth: 0
Growth Rate: 0.00%
Updated:June 11 2024
huggingface.co

Total runs: 0
Run Growth: 0
Growth Rate: 0.00%
Updated:March 12 2026
huggingface.co

Total runs: 0
Run Growth: 0
Growth Rate: 0.00%
Updated:October 24 2025
huggingface.co

Total runs: 0
Run Growth: 0
Growth Rate: 0.00%
Updated:January 31 2026
huggingface.co

Total runs: 0
Run Growth: 0
Growth Rate: 0.00%
Updated:January 12 2026
huggingface.co

Total runs: 0
Run Growth: 0
Growth Rate: 0.00%
Updated:September 01 2024
huggingface.co

Total runs: 0
Run Growth: 0
Growth Rate: 0.00%
Updated:August 19 2025
huggingface.co

Total runs: 0
Run Growth: 0
Growth Rate: 0.00%
Updated:July 23 2024