What are the hardware requirements for Qwen3 Coder 480B A35B Instruct FP8?

Qwen3 Coder 480B A35B Instruct FP8 requires approximately 429.8 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 Coder 480B A35B Instruct FP8 and how large is it?

Qwen3 Coder 480B A35B Instruct FP8 is developed by Qwen, a model with 480b 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 Coder 480B A35B Instruct FP8?

Qwen3 Coder 480B A35B Instruct FP8 is hosted on Hugging Face and linked from this page. LLM Explorer also lists quantized variants and similar alternatives if available.

Qwen3 Coder 480B A35B Instruct FP8 by Qwen — VRAM 429.8GB, 256K context

Name: Qwen3 Coder 480B A35B Instruct FP8
Author: Qwen

Qwen3 Coder 480B A35B Instruct FP8 is an open-source language model by Qwen. Features: 480b LLM, VRAM: 429.8GB, Context: 256K, License: apache-2.0, Instruction-Based, Code Generating, LLM Explorer Score: 0.29.

Arxiv:2505.09388 Codegen Conversational Deploy:azure Endpoints compatible Fp8 Instruct Qwen3 moe Region:us Safetensors Sharded Tensorflow

Model Card on HF 🤗: https://huggingface.co/Qwen/Qwen3-Coder-480B-A35B-Instruct-FP8

Qwen3 Coder 480B A35B Instruct FP8 Parameters and Internals

LLM Name	Qwen3 Coder 480B A35B Instruct FP8
Repository 🤗	https://huggingface.co/Qwen/Qwen3-Coder-480B-A35B-Instruct-FP8
Model Size	480b
Required VRAM	429.8 GB
Updated	2026-08-05
Maintainer	Qwen
Model Type	qwen3_moe
Instruction-Based	Yes
Model Files	10.0 GB: 1-of-49 10.0 GB: 2-of-49 10.0 GB: 3-of-49 10.0 GB: 4-of-49 10.0 GB: 5-of-49 10.0 GB: 6-of-49 10.0 GB: 7-of-49 10.0 GB: 8-of-49 10.0 GB: 9-of-49 10.0 GB: 10-of-49 9.9 GB: 11-of-49 10.0 GB: 12-of-49 10.0 GB: 13-of-49 10.0 GB: 14-of-49 10.0 GB: 15-of-49 10.0 GB: 16-of-49 10.0 GB: 17-of-49 10.0 GB: 18-of-49 10.0 GB: 19-of-49 10.0 GB: 20-of-49 10.0 GB: 21-of-49 10.0 GB: 22-of-49 10.0 GB: 23-of-49 10.0 GB: 24-of-49 10.0 GB: 25-of-49 10.0 GB: 26-of-49 10.0 GB: 27-of-49 9.9 GB: 28-of-49 10.0 GB: 29-of-49 10.0 GB: 30-of-49 10.0 GB: 31-of-49 10.0 GB: 32-of-49 10.0 GB: 33-of-49 10.0 GB: 34-of-49 10.0 GB: 35-of-49 10.0 GB: 36-of-49 10.0 GB: 37-of-49 10.0 GB: 38-of-49 10.0 GB: 39-of-49 10.0 GB: 40-of-49 10.0 GB: 41-of-49 10.0 GB: 42-of-49 10.0 GB: 43-of-49
Generates Code	Yes
Model Architecture	Qwen3MoeForCausalLM
License	apache-2.0
Context Length	262144
Model Max Length	262144
Transformers Version	4.52.4
Vocabulary Size	151936
Torch Data Type	bfloat16

Rank the Qwen3 Coder 480B A35B Instruct 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 Coder 480B A35B Instruct FP8

Best Alternatives	Context / RAM	Downloads	Likes
...n3 Coder 480B A35B Instruct 1M	1024K / 171.9 GB	20	8
Qwen3 Coder 480B A35B Instruct	256K / 167.9 GB	35	5
Qwen3 Coder 480B A35B Instruct	256K / 171.9 GB	29491	1356
...Coder 480B A35B Instruct NVFP4	256K / 210 GB	768	12
Qwen3 Coder REAP 363B A35B	256K / 215 GB	202	5
...3 Coder 480B A35B Instruct FP8	256K / 429.8 GB	59	8
... Coder 480B A35B Instruct 4bit	256K / 185.6 GB	2835	19

Note: green Score (e.g. "73.2") means that the model is better than Qwen/Qwen3-Coder-480B-A35B-Instruct-FP8.