Ring Lite Linear Preview is an open-source language model by inclusionAI. Features: 17.1b LLM, VRAM: 34.2GB, Context: 16K, License: mit, MoE, LLM Explorer Score: 0.18.
| LLM Name | Ring Lite Linear Preview |
| Repository 🤗 | https://huggingface.co/inclusionAI/Ring-lite-linear-preview |
| Base Model(s) | |
| Model Size | 17.1b |
| Required VRAM | 34.2 GB |
| Updated | 2026-08-08 |
| Maintainer | inclusionAI |
| Model Type | bailing_moe_linear |
| Model Files | |
| Supported Languages | zh en |
| Model Architecture | BailingMoeLinearForCausalLM |
| License | mit |
| Context Length | 16384 |
| Model Max Length | 16384 |
| Transformers Version | 4.36.0 |
| Tokenizer Class | PreTrainedTokenizerFast |
| Padding Token | <|endoftext|> |
| Vocabulary Size | 126464 |
| Torch Data Type | bfloat16 |