Latest Content
GLM 5 FP8 icon

GLM 5 FP8

NVIDIA
GLM-5-FP8 is a large Mixture-of-Experts (MoE) language model developed by Z.ai for complex reasoning, coding, and long-horizon agentic workflows. The model features a 744B parameter architecture with 40B activated parameters, built on a 78-layer transformer with 64 attention heads and a 6,144 hidden size, utilizing 256 routed experts with 8 experts activated per token. It integrates DeepSeek Sparse Attention (DSA) to reduce deployment cost while maintaining performance, and supports a ~202K token context window for large-scale multi-step reasoning tasks.
TypeMoE LLM
CapabilitiesText Generation, Instruction Following, Reasoning, Mathematical Reasoning+5 more
Group Release DateFebruary 11, 2026
Links
Paper|Blog|HF Model Card
LicenseMIT

Inference Instructions

Deploy and run this model on NVIDIA B200 GPUs using the command below. Copy the command to get started with inference.

CONSOLE
docker run -it --rm 
 --runtime=nvidia 
 --gpus all 
 --ipc=host 
 --shm-size=128g 
 -p 8000:8000 
 -v ~/.cache/huggingface:/root/.cache/huggingface 
 -e VLLM_FLOAT32_MATMUL_PRECISION=high 
 -e VLLM_USE_FLASHINFER_MOE_FP8=1 
 -e HF_TOKEN='YOUR_HF_TOKEN' 
 -e LD_LIBRARY_PATH='/usr/local/nvidia/lib64:/usr/local/nvidia/lib:/usr/lib/x86_64-linux-gnu' 
 vllm/vllm-openai:glm5 
 zai-org/GLM-5-FP8 
  --tensor-parallel-size 8 
 --max-model-len auto 
  --gpu-memory-utilization 0.90 
 --tool-call-parser glm47 
 --reasoning-parser glm45 
 --enable-auto-tool-choice 
 --max-num-seqs 1024 
 --disable-log-requests 
 --trust-remote-code
Note

Due to architectural limitations involving head size (576) and Sparse MLA, the --kv-cache-dtype fp8 flag is currently unsupported as it lacks a valid attention backend for this model configuration.

Model Benchmarks

Each model was tested with a fixed input size and total token volume while increasing concurrency to measure serving performance under load.

ITL vs Concurrency

Time to First Token

Throughput Scaling

Total Tokens/sec vs Avg TTFT

Vultr Cloud GPU

NVIDIA HGX B200

Deploy NVIDIA B200 on Vultr Cloud GPU

Deploy NVIDIA B200 instancesContact Sales