UltraLink LM by R0k1e

 »  All LLMs  »  R0k1e  »  UltraLink LM   URL Share it on

UltraLink LM is an open-source language model by R0k1e. Features: 13b LLM, VRAM: 52.3GB, Context: 4K, License: mit, Instruction-Based, LLM Explorer Score: 0.12.

  Arxiv:2402.04588 Dataset:ise-uiuc/magicoder-evo... Dataset:ise-uiuc/magicoder-oss...   Dataset:r0k1e/ultralink   Dataset:stingning/ultrachat   Endpoints compatible   Eng   Fra   Instruct   Llama   Region:us   Rus   Safetensors   Sharded   Spa   Tensorflow   Zho

UltraLink LM Benchmarks

nn.n% — How the model compares to the reference models: Anthropic Sonnet 3.5 ("so35"), GPT-4o ("gpt4o") or GPT-4 ("gpt4").

UltraLink LM Parameters and Internals

Model Type 
Transformers, Autoregressive, Multilingual
Additional Notes 
The UltraLink-LM is trained using a massive multilingual dataset, focusing on code, math, and chat abilities.
Supported Languages 
eng (high), fra (high), rus (high), spa (high), zho (high)
Training Details 
Data Sources:
UltraLink, UltraChat, Magicoder-Evol, Magicoder-OSS, MetaMathQA, ShareGPT
Hardware Used:
NVIDIA A100 80GB PCIe
Model Architecture:
Transformer style autoregressive massively multilingual language model
Input Output 
Accepted Modalities:
text
LLM NameUltraLink LM
Repository 🤗https://huggingface.co/R0k1e/UltraLink-LM 
Model Size13b
Required VRAM52.3 GB
Updated2026-04-21
MaintainerR0k1e
Model Typellama
Instruction-BasedYes
Model Files  4.9 GB: 1-of-11   5.0 GB: 2-of-11   5.0 GB: 3-of-11   5.0 GB: 4-of-11   5.0 GB: 5-of-11   4.8 GB: 6-of-11   4.8 GB: 7-of-11   4.8 GB: 8-of-11   5.0 GB: 9-of-11   5.0 GB: 10-of-11   3.0 GB: 11-of-11
Supported Languagesen
Model ArchitectureLlamaForCausalLM
Licensemit
Context Length4096
Model Max Length4096
Transformers Version4.36.2
Tokenizer ClassLlamaTokenizer
Vocabulary Size32000
Torch Data Typefloat32

Best Alternatives to UltraLink LM

Best Alternatives
Context / RAM
Downloads
Likes
Reverse Instruct32K / 27 GB123
Etri Ones Solar4K / 42.9 GB8750
Phi 3 Orpo V8.164K / 7.6 GB270
Tinyllama Python4K / 2.2 GB62
Small Instruct4K / 2.9 GB91
Taiwan LLaMa V1.04K / 26 GB3578
Taiwan LLaMa V0.04K / 26 GB81
Taiwan LLaMa V0.94K / 26 GB70
Wongwian Micro Instruct2K / 0.5 GB12190
Llama2 Ja Small Instruct2K / 1.7 GB81
Note: green Score (e.g. "73.2") means that the model is better than R0k1e/UltraLink-LM.

Rank the UltraLink LM Capabilities

🆘 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! 🌟

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? 53999 in total.

Our Social Media →  
Original data from HuggingFace, OpenCompass and various public git repos.
Check out Ag3ntum — our secure, self-hosted AI agent for server management.
Release v20260328a