LLM EXPLORER
 LLM Explorer: A Curated Large Language Model Directory and Analytics
Dark Theme
  Agents   Benchmarks   LLM  List    LLM Hosting    LLM Leaderboards    Blog    Newsfeed    Advertise

Infinity Instruct 3M 0625 Qwen2 7B by BAAI

 »  All LLMs  »  BAAI  »  Infinity Instruct 3M 0625 Qwen2 7B   URL Share it on

  89   

Infinity Instruct 3M 0625 Qwen2 7B is an open-source language model by BAAI. Features: 7b LLM, VRAM: 15.2GB, Context: 128K, License: apache-2.0, Instruction-Based, LLM Explorer Score: 0.22.

  Conversational   Dataset:baai/infinity-instruct   Deploy:azure   En   Endpoints compatible   Instruct   Qwen2   Region:us   Safetensors   Sharded   Tensorflow
Model Card on HF 🤗: https://huggingface.co/BAAI/Infinity-Instruct-3M-0625-Qwen2-7B 

Infinity Instruct 3M 0625 Qwen2 7B Benchmarks

MMLU Pro: 32.89
GPQA: 8.39
MUSR: 6.46
BBH: 34.66
IFEval: 55.54 vs 88 (so35)-36.9%
MATH Lvl 5: 19.26
LLME Score: 0.22059
nn.n% — How the model compares to the reference models: Anthropic Sonnet 3.5 ("so35"), GPT-4o ("gpt4o") or GPT-4 ("gpt4").
Infinity Instruct 3M 0625 Qwen2 7B (BAAI/Infinity-Instruct-3M-0625-Qwen2-7B)
🌟 Advertise your project 🚀

Infinity Instruct 3M 0625 Qwen2 7B Parameters and Internals

Model Type 
supervised instruction tuning, chatbot
Use Cases 
Areas:
academic research
Primary Use Cases:
chat applications with instruction tuning
Limitations:
Accuracy and reliability may vary due to the influence of uncontrollable variables
Considerations:
The accuracy of the output cannot be guaranteed. Legal liability for content not accepted.
Additional Notes 
Efforts focus on risk assessment and data generation with a target of 10M instructions finished by late June.
Supported Languages 
en (Proficient)
Training Details 
Data Sources:
BAAI/Infinity-Instruct
Data Volume:
10 million instructions (planned)
Methodology:
Supervised instruction tuning without RLHF
Hardware Used:
FlagScale acceleration techniques
Input Output 
Input Format:
Chat template sequence
Accepted Modalities:
text
Output Format:
Textual responses
Performance Tips:
Concatenate multiple training samples to optimize performance and use FlagScale for diverse acceleration techniques.
Release Notes 
Version:
0625
Date:
2024-07-09
Notes:
Model weights released for InfInstruct-Mistral-7B 0625, InfInstruct-Qwen2-7B 0625, and others.
Version:
0613
Date:
2024-06-28
Notes:
Model weight for InfInstruct-Llama3-70B 0613 released; favorable results on AlpacaEval 2.0.
LLM NameInfinity Instruct 3M 0625 Qwen2 7B
Repository 🤗https://huggingface.co/BAAI/Infinity-Instruct-3M-0625-Qwen2-7B 
Model Size7b
Required VRAM15.2 GB
Updated2026-04-06
MaintainerBAAI
Model Typeqwen2
Instruction-BasedYes
Model Files  4.9 GB: 1-of-4   4.9 GB: 2-of-4   4.3 GB: 3-of-4   1.1 GB: 4-of-4
Supported Languagesen
Model ArchitectureQwen2ForCausalLM
Licenseapache-2.0
Context Length131072
Model Max Length131072
Transformers Version4.37.2
Tokenizer ClassQwen2Tokenizer
Padding Token<|endoftext|>
Vocabulary Size152064
Torch Data Typebfloat16
Errorsreplace

Best Alternatives to Infinity Instruct 3M 0625 Qwen2 7B

Best Alternatives
Context / RAM
Downloads
Likes
Qwen2.5 7B Instruct 1M986K / 15.4 GB101838363
Hush Qwen2.5 7B RP V1.4 1M986K / 15.2 GB42
Qwen2.5 7B RRP 1M986K / 15.2 GB46
Qwen2.5 7B CelestialHarmony 1M986K / 14.8 GB37
Q2.5 Instruct 1M Harmony986K / 15.2 GB42
Impish QWEN 7B 1M986K / 15.2 GB76
COCO 7B Instruct 1M986K / 15.2 GB184
Apollo 7B986K / 15.2 GB60
Graph R1 7B986K / 15.3 GB53
Qwen2.5 7B Sky R1 Mini986K / 15.2 GB70
Note: green Score (e.g. "73.2") means that the model is better than BAAI/Infinity-Instruct-3M-0625-Qwen2-7B.

Rank the Infinity Instruct 3M 0625 Qwen2 7B 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? 52721 in total.

Email us: info@extractum.io. Our Privacy Policy  | Terms and Conditions  | Suggest an improvement.
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
  Support LLM Explorer