What are the hardware requirements for UltraLong Thinking?

UltraLong Thinking requires approximately 16.1 GB of VRAM and supports a context window of 4192K tokens. Quantized variants may run on less VRAM; see the Quantized Models section on this page.

Who developed UltraLong Thinking and how large is it?

UltraLong Thinking is developed by mergekit-community, a model with 8b 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 UltraLong Thinking?

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

UltraLong Thinking by mergekit-community — VRAM 16.1GB, 4192K context

Name: UltraLong Thinking
Author: mergekit-community

UltraLong Thinking is an open-source language model by mergekit-community. Features: 8b LLM, VRAM: 16.1GB, Context: 4192K, Instruction-Based, Merged, LLM Explorer Score: 0.17.

Merged Model Base model:mobiuslabsgmbh/deep... Base model:nvidia/llama-3.1-ne... Conversational Endpoints compatible Instruct Llama Region:us Safetensors Sharded Tensorflow

Model Card on HF 🤗: https://huggingface.co/mergekit-community/UltraLong-Thinking

UltraLong Thinking Parameters and Internals

LLM Name	UltraLong Thinking
Repository 🤗	https://huggingface.co/mergekit-community/UltraLong-Thinking
Base Model(s)	mobiuslabsgmbh/DeepSeek-R1-ReDistill-Llama3-8B-v1.1 nvidia/Llama-3.1-Nemotron-8B-UltraLong-4M-Instruct mobiuslabsgmbh/DeepSeek-R1-ReDistill-Llama3-8B-v1.1 nvidia/Llama-3.1-Nemotron-8B-UltraLong-4M-Instruct
Merged Model	Yes
Model Size	8b
Required VRAM	16.1 GB
Updated	2026-07-13
Maintainer	mergekit-community
Model Type	llama
Instruction-Based	Yes
Model Files	5.0 GB: 1-of-4 5.0 GB: 2-of-4 4.9 GB: 3-of-4 1.2 GB: 4-of-4
Model Architecture	LlamaForCausalLM
Context Length	4292608
Model Max Length	4292608
Transformers Version	4.51.1
Tokenizer Class	PreTrainedTokenizer
Vocabulary Size	129024
Torch Data Type	bfloat16

Rank the UltraLong Thinking 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 UltraLong Thinking

Best Alternatives	Context / RAM	Downloads	Likes
...otron 8B UltraLong 4M Instruct	4192K / 32.1 GB	1135	125
...a 3.1 8B UltraLong 4M Instruct	4192K / 32.1 GB	176	24
...a 3.1 8B UltraLong 2M Instruct	2096K / 32.1 GB	875	9
...otron 8B UltraLong 2M Instruct	2096K / 32.1 GB	124	18
Cthulhu 8B V1.4	1048K / 16.1 GB	10	10
...raLong 1M Instruct Abliterated	1048K / 32.1 GB	4	9
...a 3.1 8B UltraLong 1M Instruct	1048K / 32.1 GB	1387	29
...otron 8B UltraLong 1M Instruct	1048K / 32.1 GB	702	59
Zero Llama 3.1 8B Beta6	1048K / 16.1 GB	7	1
...dger Nu Llama 3.1 8B UltraLong	1048K / 16.2 GB	15	3

Note: green Score (e.g. "73.2") means that the model is better than mergekit-community/UltraLong-Thinking.