LLM EXPLORER 60,587 MODELS INDEXED

GLM 5.3 Flash NVFP4 B200 Hybrid by tiyuvta

By tiyuvta · 550 downloads
HF Repo ↗ Copy URL

GLM 5.3 Flash NVFP4 B200 Hybrid is an open-source language model by tiyuvta. Features: 184.3b LLM, VRAM: 189.1GB, License: mit, LLM Explorer Score: 0.33.

  Arxiv:2602.15763   4-bit   8-bit   B200 Base model:quantized:zai-org/g... Base model:zai-org/glm-5.3-fla...   Blackwell   Conversational   En   Fp4   Fp8   Glm5 next   Hybrid   Memra   Modelopt   Moe   Nvfp4   Quantized   Region:us   Safetensors   Sharded   Tensorflow   Tool-calling   W4a4   Zh
Model Card on HF 🤗: https://huggingface.co/tiyuvta/GLM-5.3-Flash-NVFP4-B200-hybrid 

GLM 5.3 Flash NVFP4 B200 Hybrid Parameters and Internals

LLM NameGLM 5.3 Flash NVFP4 B200 Hybrid
Repository 🤗https://huggingface.co/tiyuvta/GLM-5.3-Flash-NVFP4-B200-hybrid 
Base Model(s)  zai-org/GLM-5.3-Flash   zai-org/GLM-5.3-Flash
Model Size184.3b
Required VRAM189.1 GB
Updated2026-09-06
Maintainertiyuvta
Model Typeglm5_next
Model Files  10.0 GB: 1-of-19   10.0 GB: 2-of-19   10.0 GB: 3-of-19   10.0 GB: 4-of-19   10.0 GB: 5-of-19   10.0 GB: 6-of-19   10.0 GB: 7-of-19   10.0 GB: 8-of-19   10.0 GB: 9-of-19   10.0 GB: 10-of-19   10.0 GB: 11-of-19   10.0 GB: 12-of-19   10.0 GB: 13-of-19   10.0 GB: 14-of-19   10.0 GB: 15-of-19   10.0 GB: 16-of-19   10.0 GB: 17-of-19   10.0 GB: 18-of-19   9.1 GB: 19-of-19
Supported Languagesen zh
Model ArchitectureGlm5NextForConditionalGeneration
Licensemit
Model Max Length1048576
Transformers Version5.16.0
Tokenizer ClassTokenizersBackend
Padding Token<|endoftext|>