LLM EXPLORER 59,516 MODELS INDEXED

Internlm2 Limarp Chat 20B by intervitens

By intervitens · 10 downloads

Internlm2 Limarp Chat 20B is an open-source language model by intervitens. Features: 20b LLM, VRAM: 39.6GB, Context: 32K, License: other, Merged, LLM Explorer Score: 0.11.

  Merged Model   Arxiv:2212.04089 Base model:finetune:interviten... Base model:intervitens/internl...   Dataset:lemonilia/limarp   En   Endpoints compatible   Llama   Region:us   Safetensors   Sharded   Tensorflow

Internlm2 Limarp Chat 20B Parameters and Internals

Model Type 
text generation
Additional Notes 
Experimental model, uses mergekit for merging models
Supported Languages 
en (native)
Training Details 
Data Sources:
lemonilia/LimaRP
Methodology:
LimaRP LoRA trained on internlm2-base-20b and merged using task arithmetic
Context Length:
8192
Input Output 
Input Format:
ChatML
Accepted Modalities:
text
LLM NameInternlm2 Limarp Chat 20B
Repository πŸ€—https://huggingface.co/intervitens/internlm2-limarp-chat-20b 
Base Model(s)  Internlm2 Base 20B Llama   intervitens/internlm2-base-20b-llama
Merged ModelYes
Model Size20b
Required VRAM39.6 GB
Updated2026-08-08
Maintainerintervitens
Model Typellama
Model Files  9.7 GB: 1-of-5   9.9 GB: 2-of-5   9.9 GB: 3-of-5   9.0 GB: 4-of-5   1.1 GB: 5-of-5
Supported Languagesen
Model ArchitectureLlamaForCausalLM
Licenseother
Context Length32768
Model Max Length32768
Transformers Version4.36.2
Is Biased0
Tokenizer ClassLlamaTokenizer
Padding Token</s>
Vocabulary Size92544
Torch Data Typebfloat16

Best Alternatives to Internlm2 Limarp Chat 20B

Best Alternatives
Context / RAM
Downloads
Likes
Internlm2 5 20B Llamafied256K / 39.9 GB75
EAGLE3 GPT Oss 20B128K / 0.7 GB5070
Stellaris Internlm2 20B R51232K / 39.8 GB123
Internlm2 Chat 20B Llama Old32K / 39.6 GB83
Internlm2 Base 20B Llama32K / 39.6 GB63
Internlm2 Base 20B Llama32K / 39.6 GB60
Deita 20B32K / 39.8 GB631
Internlm2 20B Llama32K / 39.6 GB8120
Bagel 20B V04 Llama32K / 39.6 GB67
Bagel DPO 20B V04 Llama32K / 39.6 GB73
Note: green Score (e.g. "73.2") means that the model is better than intervitens/internlm2-limarp-chat-20b.