What are the hardware requirements for Inkling Small AWQ INT4?

Inkling Small AWQ INT4 requires approximately 163 GB of VRAM. Quantized variants may run on less VRAM; see the Quantized Models section on this page.

Who developed Inkling Small AWQ INT4 and how large is it?

Inkling Small AWQ INT4 is developed by cyankiwi, a model with 266b 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 Inkling Small AWQ INT4?

Inkling Small AWQ INT4 is hosted on Hugging Face and linked from this page. LLM Explorer also lists quantized variants and similar alternatives if available.

Inkling Small AWQ INT4 by cyankiwi — VRAM 163GB

Name: Inkling Small AWQ INT4
Author: cyankiwi

Inkling Small AWQ INT4 is an open-source language model by cyankiwi. Features: 266b LLM, VRAM: 163GB, License: apache-2.0, Quantized.

Audio-text-to-text Awq Base model:quantized:thinkingm... Base model:thinkingmachines/in... Compressed-tensors Conversational En Endpoints compatible Image-text-to-text Inkling mm model Moe Quantized Region:us Safetensors Sharded Tensorflow

Model Card on HF 🤗: https://huggingface.co/cyankiwi/Inkling-Small-AWQ-INT4

Inkling Small AWQ INT4 Parameters and Internals

LLM Name	Inkling Small AWQ INT4
Repository 🤗	https://huggingface.co/cyankiwi/Inkling-Small-AWQ-INT4
Base Model(s)	Inkling Small thinkingmachines/Inkling-Small
Model Size	266b
Required VRAM	163 GB
Updated	2026-08-09
Maintainer	cyankiwi
Model Type	inkling_mm_model
Model Files	4.5 GB: 1-of-37 4.5 GB: 2-of-37 4.5 GB: 3-of-37 4.5 GB: 4-of-37 4.5 GB: 5-of-37 4.5 GB: 6-of-37 4.5 GB: 7-of-37 4.5 GB: 8-of-37 4.5 GB: 9-of-37 4.5 GB: 10-of-37 4.5 GB: 11-of-37 4.5 GB: 12-of-37 4.5 GB: 13-of-37 4.5 GB: 14-of-37 4.5 GB: 15-of-37 4.5 GB: 16-of-37 4.5 GB: 17-of-37 4.5 GB: 18-of-37 4.5 GB: 19-of-37 4.5 GB: 20-of-37 4.5 GB: 21-of-37 4.5 GB: 22-of-37 4.5 GB: 23-of-37 4.5 GB: 24-of-37 4.5 GB: 25-of-37 4.5 GB: 26-of-37 4.5 GB: 27-of-37 4.5 GB: 28-of-37 4.5 GB: 29-of-37 4.5 GB: 30-of-37 4.5 GB: 31-of-37 4.5 GB: 32-of-37 4.5 GB: 33-of-37 4.5 GB: 34-of-37 4.5 GB: 35-of-37 4.5 GB: 36-of-37 1.0 GB: 37-of-37
Supported Languages	en
AWQ Quantization	Yes
Quantization Type	awq
Model Architecture	InklingForConditionalGeneration
License	apache-2.0
Tokenizer Class	PreTrainedTokenizerFast

Rank the Inkling Small AWQ INT4 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 Inkling Small AWQ INT4

Best Alternatives	Context / RAM	Downloads	Likes
Inkling Small	0K / 527.2 GB	28178	343
Inkling Small FP8 Dynamic	0K / 267.8 GB	1369	1

Note: green Score (e.g. "73.2") means that the model is better than cyankiwi/Inkling-Small-AWQ-INT4.