AI Model
Qwen3 235B A22B
235B parameters · text-generation
VRAM (FP16)
471 GB
VRAM (INT4)
118 GB
Family
qwen
Compatible GPUs
NVIDIA B300
Min GPUs: 2 · fp16
NVIDIA B200
Min GPUs: 3 · fp16
NVIDIA GB200
Min GPUs: 3 · fp16
AMD Instinct MI300X
Min GPUs: 3 · fp16
AMD Instinct MI250X
Min GPUs: 4 · fp16
NVIDIA H200
Min GPUs: 4 · fp16
NVIDIA H200 NVL
Min GPUs: 4 · fp16
NVIDIA GH200
Min GPUs: 5 · fp16
NVIDIA H100
Min GPUs: 6 · fp16
NVIDIA A100
Min GPUs: 6 · fp16
NVIDIA A16
Min GPUs: 8 · fp16
NVIDIA L40S
Min GPUs: 10 · fp16
NVIDIA A40
Min GPUs: 10 · fp16
NVIDIA RTX 6000 Ada
Min GPUs: 10 · fp16
NVIDIA RTX A6000
Min GPUs: 10 · fp16
NVIDIA V100
Min GPUs: 15 · fp16
NVIDIA RTX 5090
Min GPUs: 15 · fp16
NVIDIA L4
Min GPUs: 20 · fp16
NVIDIA A10
Min GPUs: 20 · fp16
NVIDIA A10G
Min GPUs: 20 · fp16
NVIDIA RTX 4090
Min GPUs: 20 · fp16
NVIDIA RTX A5000
Min GPUs: 20 · fp16
NVIDIA RTX 3090
Min GPUs: 20 · fp16
NVIDIA T4
Min GPUs: 30 · fp16
NVIDIA P100
Min GPUs: 30 · fp16
NVIDIA RTX 5080
Min GPUs: 30 · fp16
NVIDIA RTX A4000
Min GPUs: 30 · fp16
Supported Frameworks
vLLMPyTorchTensorRT-LLMText Generation InferenceOllamaSGLangllama.cpp
Deploy Qwen3 235B A22B
Get a full deployment stack recommendation — GPU, count, framework, quantization, and projected cost.
Start deploymentHugging Face
View the model card, tokenizer, and weights on the Hugging Face Hub.
Open on Hugging FaceVRAM Usage
FP16 serving needs about 471 GB before workload-specific headroom. INT4 quantization reduces the model weights to about 118 GB, which is the practical path for large models on smaller GPU clusters.
Related Qwen3 235B A22B resources
Move from model requirements into compatible GPU prices, deployment, and the wider model catalog.
NVIDIA B300 prices for Qwen3 235B A22BRecommended GPU path for this model at fp16 precision.NVIDIA B200 cloud pricesCompatible option for Qwen3 235B A22B; minimum 3 GPUs.NVIDIA GB200 cloud pricesCompatible option for Qwen3 235B A22B; minimum 3 GPUs.AMD Instinct MI300X cloud pricesCompatible option for Qwen3 235B A22B; minimum 3 GPUs.Deploy Qwen3 235B A22BGenerate a deployment recommendation with GPU count, framework, and estimated cost.Model VRAM leaderboardCompare FP16 and INT4 memory requirements across other deployable models.