What are the hardware requirements for Fietje 2 Chat?

Fietje 2 Chat requires approximately 5.6 GB of VRAM and supports a context window of 2K tokens. Quantized variants may run on less VRAM; see the Quantized Models section on this page.

Who developed Fietje 2 Chat and how large is it?

Fietje 2 Chat is developed by BramVanroy, a model with 2b 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 Fietje 2 Chat?

Fietje 2 Chat is hosted on Hugging Face and linked from this page. LLM Explorer also lists quantized variants and similar alternatives if available.

Fietje 2 Chat by BramVanroy — VRAM 5.6GB, 2K context

Name: Fietje 2 Chat
Author: BramVanroy

Fietje 2 Chat is an open-source language model by BramVanroy. Features: 2b LLM, VRAM: 5.6GB, Context: 2K, License: mit, Instruction-Based, LLM Explorer Score: 0.15.

Alignment-handbook Base model:bramvanroy/fietje-2... Base model:finetune:bramvanroy... Conversational Dataset:bramvanroy/orca dpo pa... Dataset:bramvanroy/ultra feedb... Dpo Fietje Instruct Nl Phi Region:us Safetensors Sharded Tensorflow Trl

Model Card on HF 🤗: https://huggingface.co/BramVanroy/fietje-2-chat

Fietje 2 Chat Parameters and Internals

Model Type

text-generation

Supported Languages

nl (proficient)

Training Details

Data Sources:

BramVanroy/ultra_feedback_dutch_cleaned, BramVanroy/orca_dpo_pairs_dutch_cleaned

Data Volume:

28 billion tokens

Methodology:

DPO-tuned continuations

Training Time:

~9 hours on one A100 80GB

Hardware Used:

Flemish Supercomputer Center (VSC)

Model Architecture:

adapted version of microsoft/phi-2

LLM Name	Fietje 2 Chat
Repository 🤗	https://huggingface.co/BramVanroy/fietje-2-chat
Base Model(s)	Fietje 2 Instruct BramVanroy/fietje-2-instruct
Model Size	2b
Required VRAM	5.6 GB
Updated	2026-08-22
Maintainer	BramVanroy
Model Type	phi
Instruction-Based	Yes
Model Files	5.0 GB: 1-of-2 0.6 GB: 2-of-2
Supported Languages	nl
Model Architecture	PhiForCausalLM
License	mit
Context Length	2048
Model Max Length	2048
Transformers Version	4.39.1
Tokenizer Class	CodeGenTokenizer
Padding Token	<\|im_end\|>
Vocabulary Size	50297
Torch Data Type	bfloat16

Rank the Fietje 2 Chat 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 Fietje 2 Chat

Best Alternatives	Context / RAM	Downloads	Likes
Fietje 2 Instruct	2K / 5.6 GB	275	6

Note: green Score (e.g. "73.2") means that the model is better than BramVanroy/fietje-2-chat.