What are the hardware requirements for Koishi Instruct 3B?

Koishi Instruct 3B requires approximately 5.7 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 Koishi Instruct 3B and how large is it?

Koishi Instruct 3B is developed by ewof, a model with 3b parameters. The model is published as open weights on Hugging Face and indexed on LLM Explorer with full benchmark history.

How does Koishi Instruct 3B perform on standard benchmarks?

Koishi Instruct 3B has the following published scores: MMLU 26.58. Compare against reference models on this page or on the LLM Explorer leaderboards.

Koishi Instruct 3B by ewof — VRAM 5.7GB, 2K context

Name: Koishi Instruct 3B
Author: ewof

Koishi Instruct 3B is an open-source language model by ewof. Features: 3b LLM, VRAM: 5.7GB, Context: 2K, Instruction-Based, LLM Explorer Score: 0.11, Arc: 41, HellaSwag: 64.5, MMLU: 26.6, GSM8K: 1.1.

Dataset:ewof/koishi-instruct-m... Endpoints compatible Gpt neox Instruct Region:us Safetensors

Model Card on HF 🤗: https://huggingface.co/ewof/koishi-instruct-3b

Koishi Instruct 3B Parameters and Internals

Training Details

Data Sources:

ewof/koishi-instruct-metharme

Input Output

Input Format:

<|system|>, <|user|>, <|model|>

LLM Name	Koishi Instruct 3B
Repository 🤗	https://huggingface.co/ewof/koishi-instruct-3b
Model Size	3b
Required VRAM	5.7 GB
Updated	2026-08-08
Maintainer	ewof
Model Type	gpt_neox
Instruction-Based	Yes
Model Files	5.7 GB
Model Architecture	GPTNeoXForCausalLM
Context Length	2048
Model Max Length	2048
Transformers Version	4.30.0.dev0
Tokenizer Class	GPTNeoXTokenizer
Vocabulary Size	50432
Torch Data Type	bfloat16

Rank the Koishi Instruct 3B 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 Koishi Instruct 3B

Best Alternatives	Context / RAM	Downloads	Likes
Stablecode Instruct Alpha 3B	4K / 6.1 GB	2	304
...ama INCITE Chat Instruct 3B V1	2K / 5.5 GB	55	1
... Instruction Tuning With GPT 4	2K / 5.7 GB	134	3
Text2table	2K / 5.7 GB	10	4
...edPajama INCITE Instruct 3B V1	2K / 5.7 GB	280	93
...blecode Instruct Alpha 3B GPTQ	4K / 1.8 GB	10	17
...blecode Instruct Alpha 3B GPTQ	4K / 1.8 GB	5	1
...ama INCITE Instruct 3B V1 GGML	2K / 1.7 GB	8	10

Note: green Score (e.g. "73.2") means that the model is better than ewof/koishi-instruct-3b.