What are the hardware requirements for Qwen3 4B Thinking 2507 Heretic CodeFeedback?

Qwen3 4B Thinking 2507 Heretic CodeFeedback requires approximately 8.1 GB of VRAM and supports a context window of 256K tokens. Quantized variants may run on less VRAM; see the Quantized Models section on this page.

Who developed Qwen3 4B Thinking 2507 Heretic CodeFeedback and how large is it?

Qwen3 4B Thinking 2507 Heretic CodeFeedback is developed by JoaoZaokk, a model with 4b parameters. The model is published as open weights on Hugging Face and indexed on LLM Explorer with full benchmark history.

Where can I download or evaluate Qwen3 4B Thinking 2507 Heretic CodeFeedback?

Qwen3 4B Thinking 2507 Heretic CodeFeedback is hosted on Hugging Face and linked from this page. LLM Explorer also lists quantized variants and similar alternatives if available.

Qwen3 4B Thinking 2507 Heretic CodeFeedback by JoaoZaokk — VRAM 8.1GB, 256K context

Name: Qwen3 4B Thinking 2507 Heretic CodeFeedback
Author: JoaoZaokk

Qwen3 4B Thinking 2507 Heretic CodeFeedback is an open-source language model by JoaoZaokk. Features: 4b LLM, VRAM: 8.1GB, Context: 256K, License: other, Instruction-Based, LLM Explorer Score: 0.24.

Base model:finetune:joaozaokk/... Base model:joaozaokk/qwen3-4b-... Code Conversational Dataset:iamtarun/python code i... Dataset:m-a-p/codefeedback-fil... En Endpoints compatible Heretic Instruct Lora Merge Python Qlora Qwen3 Region:us Safetensors Sharded Tensorflow

Model Card on HF 🤗: https://huggingface.co/JoaoZaokk/Qwen3-4B-Thinking-2507-Heretic-CodeFeedback

Qwen3 4B Thinking 2507 Heretic CodeFeedback Parameters and Internals

LLM Name	Qwen3 4B Thinking 2507 Heretic CodeFeedback
Repository 🤗	https://huggingface.co/JoaoZaokk/Qwen3-4B-Thinking-2507-Heretic-CodeFeedback
Base Model(s)	JoaoZaokk/Qwen3-4B-Thinking-2507-MiniMax-M2.1-Distill-heretic JoaoZaokk/Qwen3-4B-Thinking-2507-MiniMax-M2.1-Distill-heretic
Model Size	4b
Required VRAM	8.1 GB
Updated	2026-06-07
Maintainer	JoaoZaokk
Model Type	qwen3
Instruction-Based	Yes
Model Files	4.0 GB: 1-of-3 4.0 GB: 2-of-3 0.1 GB: 3-of-3
Supported Languages	en
Model Architecture	Qwen3ForCausalLM
License	other
Context Length	262144
Model Max Length	262144
Transformers Version	5.10.2
Tokenizer Class	Qwen2Tokenizer
Padding Token	<\|vision_pad\|>
Vocabulary Size	151936
Errors	replace

Rank the Qwen3 4B Thinking 2507 Heretic CodeFeedback capabilities

Have you tried this model? Rate its performance — 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 Qwen3 4B Thinking 2507 Heretic CodeFeedback

Best Alternatives	Context / RAM	Downloads	Likes
Qwen3 4B Instruct 2507	256K / 8.1 GB	3109972	911
FastContext 1.0 4B SFT	256K / 8.1 GB	5735	357
Fable Traces	256K / 8.1 GB	342	210
FastContext 1.0 4B RL	256K / 8.1 GB	4559	61
Qwen3 4B Cuad Extractor	256K / 8.1 GB	322	1
Qwen3 4B Instruct 2507 FP8	256K / 5.2 GB	1100744	78
Qwen Fine Tuned	256K / 8 GB	293	1
Qwen3 4B Sft Prm	256K / 17.6 GB	602	0
...pt Full Qwen3 4b Instruct 2507	256K / 8 GB	365	1
Luck Qwen3 4B Code FineTune	256K / 8.1 GB	132	1

Note: green Score (e.g. "73.2") means that the model is better than JoaoZaokk/Qwen3-4B-Thinking-2507-Heretic-CodeFeedback.