LLM

What GPU do I need to run Qwen/Qwen3-8B?

A 8.2B language model for chat and instruction-following. 8.2B parameters, published in BF16. View on Hugging Face

8.2B
Parameters
BF16
Native precision
40K tokens (40,960)
Context length
Apache 2.0
License
Text
Modality
Alibaba (Qwen)
Organization

Qwen3-8B is published by Qwen on Hugging Face, with 13,632,574 downloads and 1,335 likes to date. It's a Qwen3ForCausalLM model built for text-generation, published natively in BF16.

What Qwen3-8B is

Qwen3-8B is a 8.2B-parameter language model published by Alibaba (Qwen) on Hugging Face. It is released under Apache 2.0.

License note: fully permissive, no gating and no usage restrictions. Facts in this section are sourced from Qwen3-8B's Hugging Face model card, not benchmarked by Aquanode.

What it's used for

  • Chat assistants
  • Instruction following
  • Synthetic data generation

VRAM required & cheapest live GPU fit

Required VRAM = weight size at each precision times a flat 1.2 overhead for activations and fragmentation. The KV-cache grows with context and is not in that factor; it is listed per model below where the architecture is published. Full formula and assumptions: methodology.

PrecisionWeight sizeRequired VRAMCheapest live fitGPUs neededEst. $/hr (full fit)
BF1615.3 GB18.3 GBRTX A50001$0.176/hr
FP8 (quantized)7.6 GB9.2 GBRTX 5060 Ti1$0.180/hr
INT4 (quantized)3.8 GB4.6 GBRTX 30701$0.088/hr

A GPU is only matched to a row if its hardware supports that precision, and the primary recommendation is always a single-GPU fit when one exists.

INT4 requires a quantized checkpoint actually published for this model, check its Hugging Face page before relying on this row.

Cheapest way to run Qwen3-8B at its published (BF16) precision: 1× RTX A5000, at $0.176/hr per GPU ($0.176/hr total). Quantizing to FP8 or INT4 (rows above) can cost less, but requires a compatible quantized checkpoint to exist for this model.

Qwen3-8B: KV cache by context length

The KV cache is the memory the attention layers hold for every token of context, on top of the weights. It is not part of the flat 1.2x overhead in the table above, grows with context length and with every concurrent request, and is why a long-context deployment needs more VRAM than the table shows.

Grouped-query attention: every layer caches keys and values for a few shared KV heads. Cached per token: 147,456 bytes at 16-bit.

ContextKV cache, one sequence
4K0.56 GB
32K4.5 GB
40K (model maximum)5.6 GB

Computed from the layer, head and window counts in the model's own configuration (Qwen/Qwen3-8B), at 16 bits per cached value. Each concurrent sequence needs its own cache, on top of the weights.

Qwen3-8B: common questions

Does Qwen3-8B fit on a 24 GB GPU?

Yes. At BF16 it needs 18.3 GB of VRAM, so a 24 GB card holds it with 5.7 GB to spare. A 16 GB card is not enough for it at BF16.

What is the least VRAM Qwen3-8B can run in?

4.6 GB, at INT4 (quantized), which fits a 6 GB card, against 18.3 GB at BF16. That row assumes an INT4 checkpoint has actually been published for this model, so check its Hugging Face page before planning around it.

Does quantizing Qwen3-8B lower the GPU bill?

Yes. At BF16 the cheapest live fit is one RTX A5000 at $0.176/hr. At INT4 (quantized) it drops to one RTX 3070 at $0.088/hr, provided a quantized checkpoint exists for it.

How to run Qwen3-8B

Run Qwen3-8B with vLLM

From Qwen/Qwen3-8B's own deployment docs.

vllm serve Qwen/Qwen3-8B --enable-reasoning --reasoning-parser deepseek_r1

Source: https://huggingface.co/Qwen/Qwen3-8B/raw/main/README.md

Run Qwen3-8B with Ollama

Verified against Ollama's own library listing.

ollama run qwen3:8b

Source: https://ollama.com/library/qwen3:8b

Run Qwen3-8B with GGUF quantizations

Prebuilt GGUF weights published at unsloth/Qwen3-8B-GGUF. Run with llama.cpp's llama-server or load the repo directly in LM Studio.

llama-server -hf unsloth/Qwen3-8B-GGUF

Source: https://huggingface.co/unsloth/Qwen3-8B-GGUF

Deploy Qwen3-8B on Aquanode

Aquanode has no one-click deploy template for Qwen3-8B; you install the inference engine yourself with the commands below. Aquanode sells GPU pods billed per second, not a hosted inference API.

  1. Launch a bare GPU pod sized to the requirement above (1× RTX A5000 or larger).
  2. Open a terminal on the pod, or save one of the commands above as a startup script so it runs automatically the first time the pod boots.
  3. Run the command and connect to the resulting endpoint.
Launch a GPU pod

Weight-to-VRAM math, the fit rules, and how live prices are normalized: full methodology.

More Qwen3 models

All 105 Qwen3 models: VRAM and GPU requirements

Lineage

Based on Qwen3-8B-Base.

Quantized and fine-tuned versions

Also published as: unsloth/Qwen3-8B. Identical copies of this model under another account.

Alternatives at this size

Other models for text-generation within about a third of Qwen3-8B's 8.2B parameters, from other model lines.

More on Qwen3-8B

Related reading: RTX A5000 pricing and specs, How much VRAM you need for LLMs, Serving LLMs with vLLM, vLLM vs TensorRT-LLM vs SGLang, and Best GPU for LLM inference.

Related on Aquanode: GPU requirements for LFM2-8B-A1B, GPU requirements for c4ai-command-r-v01, and GPU requirements for llm-jp-4-33b-thinking.

Your own GPU cloud accounts

One console for the GPU clouds you already rent from. Your keys stay on your machine.

$20 a month, flat

Connect your accounts

One console for the GPU clouds you already use.

Connect your accounts for $20 a month, flat. Your provider bills the GPU.

© 2026 Aquanode. All rights reserved.

All trademarks, logos and brand names are the property of their respective owners.