What are the hardware requirements for Qwen3 4B Sky High Hermes?

Qwen3 4B Sky High Hermes requires approximately 8.2 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 Sky High Hermes and how large is it?

Qwen3 4B Sky High Hermes is developed by ZeroXClem, 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 Sky High Hermes?

Qwen3 4B Sky High Hermes is hosted on Hugging Face and linked from this page. LLM Explorer also lists quantized variants and similar alternatives if available.

Qwen3 4B Sky High Hermes by ZeroXClem — VRAM 8.2GB, 256K context

Name: Qwen3 4B Sky High Hermes
Author: ZeroXClem

Qwen3 4B Sky High Hermes is an open-source language model by ZeroXClem. Features: 4b LLM, VRAM: 8.2GB, Context: 256K, License: apache-2.0, Instruction-Based, Merged, LLM Explorer Score: 0.21.

Merged Model Abliterated Base model:davidau/qwen3-4b-in... Base model:davidau/qwen3-4b-th... Base model:qwen/qwen3-4b-think... Base model:teichai/qwen3-4b-in... Base model:teichai/qwen3-4b-th... Base model:teichai/qwen3-4b-th... Base model:teichai/qwen3-4b-th... Base model:zeroxclem/qwen3-4b-... Claude Distilled Heretic Hermes Highreasoning Instruct Qwen3 Region:us Safetensors Sharded Skyhighhermes Tensorflow Zeroxclem

Model Card on HF 🤗: https://huggingface.co/ZeroXClem/Qwen3-4B-Sky-High-Hermes

Qwen3 4B Sky High Hermes Parameters and Internals

LLM Name	Qwen3 4B Sky High Hermes
Repository 🤗	https://huggingface.co/ZeroXClem/Qwen3-4B-Sky-High-Hermes
Base Model(s)	...ha Distill Heretic Abliterated ...ng Distill Heretic Abliterated TeichAI/Qwen3-4B-Thinking-2507-Claude-Haiku-4.5-High-Reasoning-Distill TeichAI/Qwen3-4B-Instruct-2507-Claude-Opus-3-Distill TeichAI/Qwen3-4B-Thinking-2507-MiMo-V2-Flash-Distill TeichAI/Qwen3-4B-Thinking-2507-MiniMax-M2.1-Distill Qwen3 4B Thinking 2507 Qwen3 4B Hermes Axion Pro DavidAU/Qwen3-4B-Instruct-2507-Polaris-Alpha-Distill-Heretic-Abliterated DavidAU/Qwen3-4B-Thinking-2507-Gemini-3-Pro-Preview-High-Reasoning-Distill-Heretic-Abliterated TeichAI/Qwen3-4B-Thinking-2507-Claude-Haiku-4.5-High-Reasoning-Distill TeichAI/Qwen3-4B-Instruct-2507-Claude-Opus-3-Distill TeichAI/Qwen3-4B-Thinking-2507-MiMo-V2-Flash-Distill TeichAI/Qwen3-4B-Thinking-2507-MiniMax-M2.1-Distill Qwen/Qwen3-4B-Thinking-2507 ZeroXClem/Qwen3-4B-Hermes-Axion-Pro
Merged Model	Yes
Model Size	4b
Required VRAM	8.2 GB
Updated	2026-07-27
Maintainer	ZeroXClem
Model Type	qwen3
Instruction-Based	Yes
Model Files	1.0 GB: 1-of-9 1.0 GB: 2-of-9 1.0 GB: 3-of-9 1.0 GB: 4-of-9 1.0 GB: 5-of-9 1.0 GB: 6-of-9 1.0 GB: 7-of-9 1.0 GB: 8-of-9 0.2 GB: 9-of-9
Model Architecture	Qwen3ForCausalLM
License	apache-2.0
Context Length	262144
Model Max Length	262144
Transformers Version	4.57.5
Tokenizer Class	Qwen2Tokenizer
Padding Token	<\|endoftext\|>
Vocabulary Size	151669
Errors	replace

Rank the Qwen3 4B Sky High Hermes 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

Quantized Models of the Qwen3 4B Sky High Hermes

Model	Likes	Downloads	VRAM
Qwen3 4B Sky High Hermes 8bit	1	63	4 GB
Qwen3 4B Sky High Hermes 6bit	1	21	3 GB

Best Alternatives to Qwen3 4B Sky High Hermes

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	207	210
Jamjuri EDGE Preview 100	256K / 8 GB	427	1
FastContext 1.0 4B RL	256K / 8.1 GB	4559	61
Qwen3 4B Instruct 2507 FP8	256K / 5.2 GB	1100744	78
Voho Saudi Chat 4B	256K / 8 GB	468	1
... History Tutor 4B Experimental	256K / 8 GB	263	1
Qwen3 4B Cuad Extractor	256K / 8.1 GB	322	1
...en3 4B Instruct 2507 Depressed	256K / 8.1 GB	215	1

Note: green Score (e.g. "73.2") means that the model is better than ZeroXClem/Qwen3-4B-Sky-High-Hermes.