LLM EXPLORER 59,278 MODELS INDEXED

MiroThinker 1.7 Mini Int4 AutoRound by Intel

By Intel · 114 downloads
HF Repo ↗ Copy URL

MiroThinker 1.7 Mini Int4 AutoRound is an open-source language model by Intel. Features: 4.6b LLM, VRAM: 17.1GB, Context: 256K, LLM Explorer Score: 0.23.

  Arxiv:2309.05516   4-bit   Auto-round Base model:miromind-ai/mirothi... Base model:quantized:miromind-...   Conversational   Qwen3 moe   Region:us   Safetensors   Sharded   Tensorflow
Model Card on HF 🤗: https://huggingface.co/Intel/MiroThinker-1.7-mini-int4-AutoRound 

MiroThinker 1.7 Mini Int4 AutoRound Parameters and Internals

LLM NameMiroThinker 1.7 Mini Int4 AutoRound
Repository 🤗https://huggingface.co/Intel/MiroThinker-1.7-mini-int4-AutoRound 
Base Model(s)  miromind-ai/MiroThinker-1.7-mini   miromind-ai/MiroThinker-1.7-mini
Model Size4.6b
Required VRAM17.1 GB
Updated2026-08-06
MaintainerIntel
Model Typeqwen3_moe
Model Files  1.1 GB: 1-of-17   1.1 GB: 2-of-17   1.1 GB: 3-of-17   1.1 GB: 4-of-17   1.1 GB: 5-of-17   1.1 GB: 6-of-17   1.1 GB: 7-of-17   1.1 GB: 8-of-17   1.1 GB: 9-of-17   1.1 GB: 10-of-17   1.1 GB: 11-of-17   1.1 GB: 12-of-17   1.1 GB: 13-of-17   1.1 GB: 14-of-17   0.5 GB: 15-of-17   0.6 GB: 16-of-17   0.6 GB: 17-of-17
Model ArchitectureQwen3MoeForCausalLM
Context Length262144
Model Max Length262144
Transformers Version4.57.6
Tokenizer ClassQwen2Tokenizer
Padding Token<|endoftext|>
Vocabulary Size151936
Errorsreplace