What are the hardware requirements for Qwen3.8 2.4T A95B FP8?

Qwen3.8 2.4T A95B FP8 requires approximately 509.1 GB of VRAM and supports a context window of 256K tokens. Quantized variants may run on less VRAM; see the Quantized Models section on this page.

Who developed Qwen3.8 2.4T A95B FP8 and how large is it?

Qwen3.8 2.4T A95B FP8 is developed by Qwen, a model with 2446.2b 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 Qwen3.8 2.4T A95B FP8?

Qwen3.8 2.4T A95B FP8 is hosted on Hugging Face and linked from this page. LLM Explorer also lists quantized variants and similar alternatives if available.

Qwen3.8 2.4T A95B FP8 by Qwen — VRAM 509.1GB, 256K context

Name: Qwen3.8 2.4T A95B FP8
Author: Qwen

Qwen3.8 2.4T A95B FP8 is an open-source language model by Qwen. Features: 2446.2b LLM, VRAM: 509.1GB, Context: 256K, License: other, MoE, LLM Explorer Score: 0.34.

Base model:quantized:qwen/qwen... Base model:qwen/qwen3.8-2.4t-a... Conversational Endpoints compatible Fp8 Moe Qwen3 5 moe text Region:us Safetensors Sharded Tensorflow

Model Card on HF 🤗: https://huggingface.co/Qwen/Qwen3.8-2.4T-A95B-FP8

Qwen3.8 2.4T A95B FP8 Parameters and Internals

LLM Name	Qwen3.8 2.4T A95B FP8
Repository 🤗	https://huggingface.co/Qwen/Qwen3.8-2.4T-A95B-FP8
Base Model(s)	Qwen/Qwen3.8-2.4T-A95B Qwen/Qwen3.8-2.4T-A95B
Model Size	2446.2b
Required VRAM	509.1 GB
Updated	2026-08-12
Maintainer	Qwen
Model Type	qwen3_5_moe_text
Model Files	17.2 GB: 1-of-213 8.6 GB: 2-of-213 17.2 GB: 3-of-213 8.6 GB: 4-of-213 17.2 GB: 5-of-213 8.6 GB: 6-of-213 17.2 GB: 7-of-213 8.6 GB: 8-of-213 17.2 GB: 9-of-213 8.6 GB: 10-of-213 4.0 GB: 11-of-213 17.2 GB: 12-of-213 8.6 GB: 13-of-213 17.2 GB: 14-of-213 8.6 GB: 15-of-213 17.2 GB: 16-of-213 8.6 GB: 17-of-213 17.2 GB: 18-of-213 8.6 GB: 19-of-213 4.0 GB: 20-of-213 17.2 GB: 21-of-213 8.6 GB: 22-of-213 17.2 GB: 23-of-213 8.6 GB: 24-of-213 4.1 GB: 25-of-213 2.8 GB: 26-of-213 17.2 GB: 27-of-213 8.6 GB: 28-of-213 17.2 GB: 29-of-213 8.6 GB: 30-of-213 17.2 GB: 31-of-213 8.6 GB: 32-of-213 17.2 GB: 33-of-213 8.6 GB: 34-of-213 17.2 GB: 35-of-213 8.6 GB: 36-of-213 4.0 GB: 37-of-213 17.2 GB: 38-of-213 8.6 GB: 39-of-213 17.2 GB: 40-of-213 8.6 GB: 41-of-213 17.2 GB: 42-of-213 8.6 GB: 43-of-213
Model Architecture	Qwen3_5MoeForCausalLM
License	other
Context Length	262144
Model Max Length	262144
Transformers Version	4.57.3
Vocabulary Size	248320

Rank the Qwen3.8 2.4T A95B FP8 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 Qwen3.8 2.4T A95B FP8

Best Alternatives	Context / RAM	Downloads	Likes
Qwen3.8 2.4T A95B	256K / 999.3 GB	978	407
Qwen3.8 2.4T A95B	256K / 999.3 GB	145	1
Qwen3.8 2.4T A95B	256K / 999.3 GB	281	1
...8 2.4T A95B MXFP4 CT AutoRound	256K / 273.4 GB	106	1

Note: green Score (e.g. "73.2") means that the model is better than Qwen/Qwen3.8-2.4T-A95B-FP8.