MiroThinker V1.5 30B by mlx-community

 ยป  All LLMs  ยป  mlx-community  ยป  MiroThinker V1.5 30B   URL Share it on

MiroThinker V1.5 30B is an open-source language model by mlx-community. Features: 30b LLM, VRAM: 17.2GB, Context: 256K, License: mit.

  4-bit   Agent Base model:miromind-ai/mirothi... Base model:quantized:miromind-...   Conversational   Deep-research   En   Endpoints compatible   Miromind   Mlx   Mlx-my-repo   Open-source   Qwen3 moe   Region:us   Safetensors   Sharded   Tensorflow
MiroThinker V1.5 30B (mlx-community/MiroThinker-v1.5-30B)
๐ŸŒŸ Advertise your project ๐Ÿš€

MiroThinker V1.5 30B Parameters and Internals

LLM NameMiroThinker V1.5 30B
Repository ๐Ÿค—https://huggingface.co/mlx-community/MiroThinker-v1.5-30B 
Base Model(s)  miromind-ai/MiroThinker-v1.5-30B   miromind-ai/MiroThinker-v1.5-30B
Model Size30b
Required VRAM17.2 GB
Updated2026-03-27
Maintainermlx-community
Model Typeqwen3_moe
Model Files  5.3 GB: 1-of-4   5.4 GB: 2-of-4   5.3 GB: 3-of-4   1.2 GB: 4-of-4
Supported Languagesen
Model ArchitectureQwen3MoeForCausalLM
Licensemit
Context Length262144
Model Max Length262144
Transformers Version4.51.1
Tokenizer ClassQwen2Tokenizer
Padding Token<|endoftext|>
Vocabulary Size151936
Torch Data Typebfloat16
Errorsreplace

Best Alternatives to MiroThinker V1.5 30B

Best Alternatives
Context / RAM
Downloads
Likes
... 42B A3B Thinking Total Recall986K / 84.8 GB260
Qwen3 30B A3B YOYO V3986K / 61.1 GB87
...07 YOYO2 TOTAL RECALL Instruct986K / 84.8 GB271
...07 YOYO2 TOTAL RECALL Instruct986K / 106 GB200
Qwen3 30B A3B YOYO V2986K / 61.1 GB85
Qwen3 Coder 30B A3B Instruct256K / 61.1 GB1045079981
Qwen3 30B A3B Instruct 2507256K / 61.1 GB1497250600
Qwen3 30B A3B Thinking 2507256K / 61.1 GB189246279
...wen3 30B A3B Instruct 2507 FP8256K / 31.2 GB13651375
...en3 Coder 30B A3B Instruct FP8256K / 31.2 GB8620691
Note: green Score (e.g. "73.2") means that the model is better than mlx-community/MiroThinker-v1.5-30B.

Rank the MiroThinker V1.5 30B Capabilities

๐Ÿ†˜ Have you tried this model? Rate its performance. This feedback would greatly assist ML community in identifying the most suitable model for their needs. Your contribution really does make a difference! ๐ŸŒŸ

Instruction Following and Task Automation  
Factuality and Completeness of Knowledge  
Censorship and Alignment  
Data Analysis and Insight Generation  
Text Generation  
Text Summarization and Feature Extraction  
Code Generation  
Multi-Language Support and Translation  

What open-source LLMs or SLMs are you in search of? 51648 in total.

Our Social Media →  
Original data from HuggingFace, OpenCompass and various public git repos.
Check out Ag3ntum โ€” our secure, self-hosted AI agent for server management.
Release v20260327b