What are the hardware requirements for Koishi 120B Qlora Gptq?

Koishi 120B Qlora Gptq requires approximately 59.8 GB of VRAM and supports a context window of 4K tokens. Quantized variants may run on less VRAM; see the Quantized Models section on this page.

Who developed Koishi 120B Qlora Gptq and how large is it?

Koishi 120B Qlora Gptq is developed by ewof, a model with 120b 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 Koishi 120B Qlora Gptq?

Koishi 120B Qlora Gptq is hosted on Hugging Face and linked from this page. LLM Explorer also lists quantized variants and similar alternatives if available.

Koishi 120B Qlora Gptq by ewof — VRAM 59.8GB, 4K context

Name: Koishi 120B Qlora Gptq
Author: ewof

Koishi 120B Qlora Gptq is an open-source language model by ewof. Features: 120b LLM, VRAM: 59.8GB, Context: 4K, Quantized, Instruction-Based, LLM Explorer Score: 0.1.

4bit Dataset:ewof/koishi-instruct-m... Endpoints compatible Gptq Instruct Llama Quantized Region:us Sharded

Model Card on HF 🤗: https://huggingface.co/ewof/koishi-120b-qlora-gptq

Koishi 120B Qlora Gptq Parameters and Internals

Training Details

Data Sources:

ewof/koishi-instruct-metharme

Methodology:

Trained on prompts using roles denoted by tokens: '<|system|>', '<|user|>', and '<|model|>'.

Context Length:

2048

Hardware Used:

8x Nvidia A100 GPU cluster

LLM Name	Koishi 120B Qlora Gptq
Repository 🤗	https://huggingface.co/ewof/koishi-120b-qlora-gptq
Base Model(s)	Koishi 120B Qlora ewof/koishi-120b-qlora
Model Size	120b
Required VRAM	59.8 GB
Updated	2026-07-31
Maintainer	ewof
Model Type	llama
Instruction-Based	Yes
Model Files	10.0 GB: 1-of-6 9.9 GB: 2-of-6 10.0 GB: 3-of-6 10.0 GB: 4-of-6 10.0 GB: 5-of-6 9.9 GB: 6-of-6
GPTQ Quantization	Yes
Quantization Type	gptq\|4bit
Model Architecture	LlamaForCausalLM
Context Length	4096
Model Max Length	4096
Transformers Version	4.35.1
Tokenizer Class	LlamaTokenizer
Padding Token	</s>
Vocabulary Size	32003
Torch Data Type	float16

Rank the Koishi 120B Qlora Gptq 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 120B Qlora Gptq

Best Alternatives	Context / RAM	Downloads	Likes
MegaDolphin 120B GPTQ	4K / 61.1 GB	25	4
...t 120B Cat A Llama EXL2 5.5bpw	8K / 85.3 GB	5	0
...t 120B Cat A Llama EXL2 4.5bpw	8K / 70.3 GB	4	1
...egaDolphin 120B 2.9bpw H6 EXL2	4K / 44.3 GB	2	3
...gaDolphin 120B 2.65bpw H6 EXL2	4K / 40.5 GB	3	2
...egaDolphin 120B 4.0bpw H6 EXL2	4K / 60.8 GB	4	1
...ma 3 Instruct 120B Cat A Llama	8K / 243.9 GB	2	1
Meta Llama 3 225B Instruct	8K / 443.2 GB	274	18
...0B Instruct Abliterated Merged	8K / 243.7 GB	1	1
MegaDolphin 120B AWQ	4K / 63.3 GB	7	2

Note: green Score (e.g. "73.2") means that the model is better than ewof/koishi-120b-qlora-gptq.