LLM EXPLORER 60,147 MODELS INDEXED

GLM 5.3 Flash JANG by JANGQ-AI

By JANGQ-AI · 247 downloads
HF Repo ↗ Copy URL

GLM 5.3 Flash JANG is an open-source language model by JANGQ-AI. Features: 28.6b LLM, VRAM: 102.4GB, License: mit, Quantized, LLM Explorer Score: 0.46, ELO: 1469.

  Agent   Apple-silicon   Awq Base model:finetune:zai-org/gl... Base model:zai-org/glm-5.3-fla...   Conversational   En   Glm5 next   Image-text-to-text   Imatrix   Jang   Mlx   Moe   Quantized   Reasoning   Region:us   Safetensors   Sharded   Tensorflow   Tool-use   Video   Vision
Model Card on HF 🤗: https://huggingface.co/JANGQ-AI/GLM-5.3-Flash-JANG 

GLM 5.3 Flash JANG Benchmarks

LMSys ELO 1469 vs 1272 (so35) 15.5%
nn.n% — How the model compares to the reference models: Anthropic Sonnet 3.5 ("so35"), GPT-4o ("gpt4o") or GPT-4 ("gpt4").

GLM 5.3 Flash JANG Parameters and Internals

LLM NameGLM 5.3 Flash JANG
Repository 🤗https://huggingface.co/JANGQ-AI/GLM-5.3-Flash-JANG 
Base Model(s)  zai-org/GLM-5.3-Flash   zai-org/GLM-5.3-Flash
Model Size28.6b
Required VRAM102.4 GB
Updated2026-09-01
MaintainerJANGQ-AI
Model Typeglm5_next
Model Files  5.9 GB: 1-of-26   0.2 GB: 2-of-26   5.8 GB: 3-of-26   3.1 GB: 4-of-26   5.8 GB: 5-of-26   3.1 GB: 6-of-26   6.1 GB: 7-of-26   1.7 GB: 8-of-26   5.4 GB: 9-of-26   3.7 GB: 10-of-26   5.4 GB: 11-of-26   3.7 GB: 12-of-26   5.6 GB: 13-of-26   4.0 GB: 14-of-26   6.3 GB: 15-of-26   3.7 GB: 16-of-26   6.0 GB: 17-of-26   4.0 GB: 18-of-26   5.4 GB: 19-of-26   4.3 GB: 20-of-26   5.4 GB: 21-of-26   2.8 GB: 22-of-26   4.5 GB: 23-of-26   0.1 GB: 24-of-26   0.2 GB: 25-of-26   0.2 GB: 26-of-26
Supported Languagesen
AWQ QuantizationYes
Quantization Typeawq
Model ArchitectureGlm5NextForConditionalGeneration
Licensemit
Model Max Length1048576
Transformers Version5.16.0
Tokenizer ClassTokenizersBackend
Padding Token<|endoftext|>