LLM EXPLORER 63,636 MODELS INDEXED

GLM 5 0.88B MTP by inference-optimization

By inference-optimization · 9 downloads

GLM 5 0.88B MTP is an open-source language model by inference-optimization. Features: 881m LLM, VRAM: 1.6GB, Context: 198K, License: mit, MoE, LLM Explorer Score: 0.26.

  Conversational   Endpoints compatible   Glm moe dsa   Llm-compressor   Moe   Mtp   Pr-3225   Region:us   Safetensors   Sharded   Tensorflow   Test-fixture   Tiny-model

GLM 5 0.88B MTP Parameters and Internals

LLM NameGLM 5 0.88B MTP
Repository πŸ€—https://huggingface.co/inference-optimization/GLM-5-0.88B-MTP 
Model Size881m
Required VRAM1.6 GB
Updated2026-10-10
Maintainerinference-optimization
Model Typeglm_moe_dsa
Model Files  0.5 GB: 1-of-4   0.5 GB: 2-of-4   0.5 GB: 3-of-4   0.1 GB: 4-of-4   0.2 GB
Model ArchitectureGlmMoeDsaForCausalLM
Licensemit
Context Length202752
Model Max Length202752
Transformers Version5.17.0
Tokenizer ClassTokenizersBackend
Padding Token<|endoftext|>
Vocabulary Size154880