What are the hardware requirements for Kestrel 4B Coder OQ4e Fp16 Mtp?

Kestrel 4B Coder OQ4e Fp16 Mtp requires approximately 3.9 GB of VRAM. Quantized variants may run on less VRAM; see the Quantized Models section on this page.

Who developed Kestrel 4B Coder OQ4e Fp16 Mtp and how large is it?

Kestrel 4B Coder OQ4e Fp16 Mtp is developed by sermonis-ai, 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 Kestrel 4B Coder OQ4e Fp16 Mtp?

Kestrel 4B Coder OQ4e Fp16 Mtp is hosted on Hugging Face and linked from this page. LLM Explorer also lists quantized variants and similar alternatives if available.

Kestrel 4B Coder OQ4e Fp16 Mtp by sermonis-ai — VRAM 3.9GB

Name: Kestrel 4B Coder OQ4e Fp16 Mtp
Author: sermonis-ai

Kestrel 4B Coder OQ4e Fp16 Mtp is an open-source language model by sermonis-ai. Features: 4b LLM, VRAM: 3.9GB, License: apache-2.0, Quantized, Code Generating, LLM Explorer Score: 0.3.

4-bit Apple-silicon Base model:jackrong/kestrel-4b... Base model:quantized:jackrong/... Code Codegen Conversational Fp16 Image-text-to-text Imatrix Mlx Mtp Multimodal Oq Oq4e Quantized Qwen3 5 Region:us Safetensors Vision

Model Card on HF 🤗: https://huggingface.co/sermonis-ai/Kestrel-4B-Coder-oQ4e-fp16-mtp

Kestrel 4B Coder OQ4e Fp16 Mtp Parameters and Internals

LLM Name	Kestrel 4B Coder OQ4e Fp16 Mtp
Repository 🤗	https://huggingface.co/sermonis-ai/Kestrel-4B-Coder-oQ4e-fp16-mtp
Model Name	unsloth/Qwen3.5-4B
Base Model(s)	Kestrel 4B Coder Jackrong/Kestrel-4B-Coder
Model Size	4b
Required VRAM	3.9 GB
Updated	2026-11-01
Maintainer	sermonis-ai
Model Type	qwen3_5
Model Files	3.9 GB
Quantization Type	fp16
Generates Code	Yes
Model Architecture	Qwen3_5ForConditionalGeneration
License	apache-2.0
Model Max Length	262144
Tokenizer Class	Qwen2Tokenizer
Padding Token	<\|vision_pad\|>
Torch Data Type	float16
Errors	replace

Rank the Kestrel 4B Coder OQ4e Fp16 Mtp 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 Kestrel 4B Coder OQ4e Fp16 Mtp

Best Alternatives	Context / RAM	Downloads	Likes
Qwen3.5 4B 3D Coder	0K / 9.1 GB	44825	465
Kestrel 4B Coder OQ4e Mtp	0K / 3.3 GB	109	1
PyRUA Lean Coder V0.1 4B	0K / 0.2 GB	48	1
Dare4b	0K / 9.1 GB	39	3
Qwopus3.5 4B Coder Fable5 V1	0K / 9.3 GB	159	1
Qwen3.5 4B Agentic Coder V4	0K / 9.3 GB	51	2

Note: green Score (e.g. "73.2") means that the model is better than sermonis-ai/Kestrel-4B-Coder-oQ4e-fp16-mtp.