What are the hardware requirements for Klear 46B A2.5B Base?

Klear 46B A2.5B Base requires approximately 92.5 GB of VRAM and supports a context window of 64K tokens. Quantized variants may run on less VRAM; see the Quantized Models section on this page.

Who developed Klear 46B A2.5B Base and how large is it?

Klear 46B A2.5B Base is developed by Kwai-Klear, a model with 46b 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 Klear 46B A2.5B Base?

Klear 46B A2.5B Base is hosted on Hugging Face and linked from this page. LLM Explorer also lists quantized variants and similar alternatives if available.

Klear 46B A2.5B Base by Kwai-Klear — VRAM 92.5GB, 64K context

Name: Klear 46B A2.5B Base
Author: Kwai-Klear

Klear 46B A2.5B Base is an open-source language model by Kwai-Klear. Features: 46b LLM, VRAM: 92.5GB, Context: 64K, License: apache-2.0, LLM Explorer Score: 0.2.

Conversational Custom code En Klear Region:us Safetensors Sharded Tensorflow Zh

Model Card on HF 🤗: https://huggingface.co/Kwai-Klear/Klear-46B-A2.5B-Base

Klear 46B A2.5B Base Parameters and Internals

LLM Name	Klear 46B A2.5B Base
Repository 🤗	https://huggingface.co/Kwai-Klear/Klear-46B-A2.5B-Base
Model Size	46b
Required VRAM	92.5 GB
Updated	2026-08-04
Maintainer	Kwai-Klear
Model Type	Klear
Model Files	5.0 GB: 1-of-19 5.0 GB: 2-of-19 5.0 GB: 3-of-19 5.0 GB: 4-of-19 5.0 GB: 5-of-19 5.0 GB: 6-of-19 5.0 GB: 7-of-19 5.0 GB: 8-of-19 5.0 GB: 9-of-19 5.0 GB: 10-of-19 5.0 GB: 11-of-19 5.0 GB: 12-of-19 5.0 GB: 13-of-19 5.0 GB: 14-of-19 5.0 GB: 15-of-19 5.0 GB: 16-of-19 5.0 GB: 17-of-19 5.0 GB: 18-of-19 2.5 GB: 19-of-19
Supported Languages	zh en
Model Architecture	KlearMoeForCausalLM
License	apache-2.0
Context Length	65536
Model Max Length	65536
Transformers Version	4.56.0
Tokenizer Class	Qwen2Tokenizer
Padding Token	<\|endoftext\|>
Vocabulary Size	151936
Errors	replace

Rank the Klear 46B A2.5B Base 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 Klear 46B A2.5B Base

Best Alternatives	Context / RAM	Downloads	Likes
Klear 46B A2.5B Instruct	64K / 92.5 GB	107	80

Note: green Score (e.g. "73.2") means that the model is better than Kwai-Klear/Klear-46B-A2.5B-Base.