Arrakis / Models / Qwen3.8-Flash-Next

Qwen3.8-Flash-Next GPU requirements

Qwen3.8-Flash-Next has 180B parameters. To serve it, plan for about 432 GB of GPU memory in 16-bit, 216 GB in 8-bit or 108 GB in 4-bit, including about 20% for the KV cache and activations.

It is a mixture of experts: 512 experts, 10 of them active for each token. All of them stay in GPU memory, so plan memory for all 180B parameters.

Deploy this model Check another model

What it takes to serve it

GPU memory to run it for inference, with the smallest GPU setup that holds it and the lowest price on Arrakis right now.

Precision	Memory and the smallest setup
16-bit	432 GB: 2 × B300, from $9.20 per hour for all 2
8-bit	216 GB: one B300, from $4.60 per GPU-hour
4-bit	108 GB: one H200, from $3.44 per GPU-hour

What it takes to fine-tune it

QLoRA trains adapters on a 4-bit frozen model, LoRA on a 16-bit one, and full fine-tuning updates every weight. Figures are before activations, which grow with batch size and sequence length.

Method	Memory and the smallest setup
QLoRA	126 GB: one H200, from $3.44 per GPU-hour
LoRA	432 GB: 2 × B300, from $9.20 per hour for all 2
Full fine-tuning	2,880 GB: more than one 8-GPU machine

About the model

From its card and files on Hugging Face. Memory figures here follow the GPU memory guide.

Parameters: 180B
Experts: 512
Experts for each token: 10
Context window (tokens): 262,144
Published in: BF16
License: other
Released: 2026-08-24
Weights: Qwen/Qwen3.8-Flash-Next

Deploy it on Arrakis

Open Models in the console, choose vLLM, SGLang or llama.cpp, and rent a GPU it fits on. You get an OpenAI-compatible endpoint with its own API key, and root SSH to the machine. Or run it yourself on any machine with enough GPU memory:

vLLM

vllm serve Qwen/Qwen3.8-Flash-Next --tensor-parallel-size 2

Deploy this model How deploying works

Questions

Live prices for every GPU are on the GPU prices page.

How much GPU memory does Qwen3.8-Flash-Next need?

To serve it, about 432 GB in 16-bit, 216 GB in 8-bit, 108 GB in 4-bit, including about 20% for the KV cache and activations. To fine-tune it, about 126 GB with QLoRA, 432 GB with LoRA and 2,880 GB for full fine-tuning with Adam, before activations.

Can I run Qwen3.8-Flash-Next on one GPU?

Yes, quantized. In 8-bit it needs about 216 GB, which fits on one B300. In 16-bit it needs 2 B300 GPUs.

What does it cost to run Qwen3.8-Flash-Next?

In 16-bit it needs 2 B300 GPUs; in 4-bit, one H200. Each setup on this page shows the lowest price per GPU-hour on Arrakis right now, and you pay by the hour from a prepaid balance.

How do I deploy Qwen3.8-Flash-Next on Arrakis?

Open Models in the Arrakis console, paste Qwen/Qwen3.8-Flash-Next, choose vLLM, SGLang or llama.cpp, and rent a GPU it fits on. You get an OpenAI-compatible endpoint with its own API key, and root SSH to the machine.

More models

Every model, and a check for any other.