LLM EXPLORER 59,182 MODELS INDEXED

Ling 3.0 Flash Base 30T by inclusionAI

By inclusionAI · 934 downloads
HF Repo ↗ Copy URL

Ling 3.0 Flash Base 30T is an open-source language model by inclusionAI. Features: 127.5b LLM, VRAM: 255.9GB, Context: 8K, License: mit, LLM Explorer Score: 0.33.

  Arxiv:2507.17634   Bailing hybrid   Custom code   Region:us   Safetensors   Sharded   Tensorflow
Model Card on HF 🤗: https://huggingface.co/inclusionAI/Ling-3.0-flash-base-30T 

Ling 3.0 Flash Base 30T Parameters and Internals

LLM NameLing 3.0 Flash Base 30T
Repository 🤗https://huggingface.co/inclusionAI/Ling-3.0-flash-base-30T 
Model Size127.5b
Required VRAM255.9 GB
Updated2026-08-21
MaintainerinclusionAI
Model Typebailing_hybrid
Model Files  10.1 GB: 1-of-27   10.1 GB: 2-of-27   10.1 GB: 3-of-27   10.1 GB: 4-of-27   8.1 GB: 5-of-27   10.1 GB: 6-of-27   10.1 GB: 7-of-27   10.1 GB: 8-of-27   10.1 GB: 9-of-27   10.1 GB: 10-of-27   8.1 GB: 11-of-27   10.1 GB: 12-of-27   8.1 GB: 13-of-27   10.1 GB: 14-of-27   8.1 GB: 15-of-27   10.1 GB: 16-of-27   10.1 GB: 17-of-27   10.1 GB: 18-of-27   10.1 GB: 19-of-27   10.1 GB: 20-of-27   10.1 GB: 21-of-27   8.1 GB: 22-of-27   10.1 GB: 23-of-27   8.9 GB: 24-of-27   10.1 GB: 25-of-27   10.7 GB: 26-of-27   3.9 GB: 27-of-27
Model ArchitectureBailingMoeV3ForCausalLM
Licensemit
Context Length8192
Model Max Length8192
Transformers Version4.56.2
Tokenizer ClassPreTrainedTokenizerFast
Padding Token<|endoftext|>
Vocabulary Size157184
Torch Data Typebfloat16

Best Alternatives to Ling 3.0 Flash Base 30T

Best Alternatives
Context / RAM
Downloads
Likes
Ling 3.0 Flash256K / 254.5 GB4189213
Ling 3.0 Flash Int4256K / 77.6 GB19699
Ling 3.0 Flash Fp8256K / 129.3 GB139526
Ling 3.0 Flash Heretic256K / 255 GB2071
...0 Flash CIRU Int4 Strix Native256K / 77.6 GB477
Ling 3.0 Flash MXFP4128K / 77.9 GB13996
Note: green Score (e.g. "73.2") means that the model is better than inclusionAI/Ling-3.0-flash-base-30T.