GLM 5
GLM-5 is a text-generation model from zai-org (repo: zai-org/GLM-5) released under the MIT license. According to its model card, it targets complex systems engineering and long-horizon agentic tasks, scales to 744B parameters (40B active) with 28.5T pre-training tokens, and integrates DeepSeek Sparse Attention. It supports English and Chinese and is intended for use with Transformers.
At a glance
- Parameters
- 753.86B
- Architecture
- glm_moe_dsa
- Context length
- 202,752
- License
- mit
- Disk space
- 1404.2 GB
- Software
- SGLang, vLLM
Will it run on your machine?
Save your machine to see a personalized rating and its reasoning.
Add my hardwareBest for
- Text generation in English and Chinese
- Complex systems engineering tasks
- Long-horizon agentic tasks
- Reasoning tasks
- Coding tasks
Tradeoffs
- Only English and Chinese are listed as supported languages
- Specific hardware requirements are not provided in the source facts
- No stated compatibility beyond the transformers library and safetensors format
- No deployment, latency, or throughput guarantees are provided
- Benchmark figures are self-reported in the model card and not independently verified
Ways to run it
SGLang · Linux
Repository-specific command in publisher documentation; confirm dependencies and hardware in the source.
sglang serve --model-path zai-org/GLM-5 --tp-size 8 --tool-call-parser glm47 --reasoning-parser glm45 --speculative-algorithm EAGLE --speculative-num-steps 3 --speculative-eagle-topk 1 --speculative-num-draft-tokens 4 --mem-fraction-static 0.85 --served-model-name glm-5vLLM · Linux
Repository-specific command in publisher documentation; confirm dependencies and hardware in the source.
vllm serve zai-org/GLM-5 --tensor-parallel-size 8 --gpu-memory-utilization 0.85 --speculative-config.method mtp --speculative-config.num_speculative_tokens 3 --tool-call-parser glm47 --reasoning-parser glm45 --enable-auto-tool-choice --served-model-name glm-5Explore its uses
Keep exploring
Bonsai 2 · 27B
A compressed 27B-class reasoning model with publisher-provided GGUF packs and a dedicated llama.cpp fork for CUDA, Metal, and CPU.
DeepSeek V4.1 Flash
A multimodal reasoning model with a compressed key-value cache, published as open weights.
Qwen3.8-Flash-Next
An experimental open-weight multimodal model with sparse attention and 262K native context.