Flow Judge V0.1 AWQ is an open-source language model by flowaicom. Features: 3.8b LLM, VRAM: 2.3GB, Context: 128K, License: apache-2.0, Quantized, Instruction-Based, Merged, LLM Explorer Score: 0.16.
Flow Judge V0.1 AWQ Parameters and Internals
Model Type
Use Cases
Primary Use Cases: Custom LLM system evaluation tasks.
Limitations: Performance in specialized task domains like arithmetic or code evaluation may be limited., Support in non-English contexts has not been rigorously tested., Handling long context inputs or parsing structured data formats like JSON is limited.
Additional Notes Model is quantized using Flow AI's AWQ safetensors quant.
Supported Languages languages_supported (English), proficiency (Full support)
Training Details
Data Sources: Synthetically generated datasets
Methodology: Supervised Fine-Tuning (SFT), RSLoRa fine-tuning, synthetic dataset generation
Context Length:
Model Architecture:
Input Output
Input Format: Text prompt format with or without user inputs
Accepted Modalities:
Output Format: Structured evaluation outputs including verbal feedback with <tags> and numeric scores.
Performance Tips: Ensure modern GPU use and adherence to prompt templates for efficient evaluations.
LLM Name Flow Judge V0.1 AWQ Repository π€ https://huggingface.co/flowaicom/Flow-Judge-v0.1-AWQ Model Name Flow-Judge-v0.1-AWQ Model Creator Flow AI Base Model(s) microsoft/Phi-3.5-mini-instruct microsoft/Phi-3.5-mini-instruct Merged Model Yes Model Size 3.8b Required VRAM 2.3 GB Updated 2026-08-08 Maintainer flowaicom Model Type phi3.5 Instruction-Based Yes Model Files 2.3 GB Supported Languages en AWQ Quantization Yes Quantization Type awq Model Architecture Phi3ForCausalLM License apache-2.0 Context Length 131072 Model Max Length 131072 Transformers Version 4.44.2 Tokenizer Class LlamaTokenizer Padding Token <|endoftext|> Vocabulary Size 32064 Torch Data Type bfloat16
Rank the Flow Judge V0.1 AWQ capabilities
Have you tried this model? Rate its performance β this feedback helps the ML community find the right model for their needs.
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
Best Alternatives to Flow Judge V0.1 AWQ
Note: green Score (e.g. "73.2 ") means that the model is better than flowaicom/Flow-Judge-v0.1-AWQ .
Expand