GPU Rental Prices
SearchCtrl K

Qwen3 Coder 480B A35B VRAM requirements

Qwen3 Coder 480B A35B has 480.15 billion parameters in its published weight files. At 4 bits, the weights alone need approximately 240.1 GB. The default inference scenario estimates 289.3 GB including cache and runtime headroom.

Checked 20 October 2026. Memory estimates use decimal GB. Quantized weights may add scales, metadata and buffers; a compatible implementation and model licence are still required.

Change the memory assumptions

Inference estimate
289.3GB
Weights 240.1 GB · cache 1.0 GB · headroom 48.2 GB.

Full inference not modelled; no GPU fit is claimed.

Memory at different precisions

PrecisionWeights GBInference GBLoRA scenario GBFull fine-tune GB
16-bit960.31,153.61,048.78,644.8
8-bit480.2577.4568.68,644.8
4-bit240.1289.3328.58,644.8

How the estimate is calculated

Weights = parameters × bits ÷ 8. For supported standard attention, KV cache = 2 × layers × KV heads × head dimension × context tokens × batch size × 2 bytes. Inference adds the chosen percentage of runtime headroom to weights and cache. Divide bytes by 1,000,000,000 for GB.

The LoRA scenario keeps the chosen base weights and allows 18 bytes per trainable adapter parameter, plus the activation allowance. Full fine-tuning uses 18 bytes per parameter for a mixed-precision AdamW scenario, plus that allowance. Its total does not fall when you select 4-bit inference weights. These are budget estimates, not measured memory peaks; training context, optimizer, checkpointing and implementation can change them substantially.

Single-GPU candidates for this inference scenario

The offers update with the memory inputs above. Availability not reported means stock has not been confirmed. Memory capacity alone does not establish runtime compatibility or speed.

More filters: provider, memory, node size, stock, place, sort, price unit

“Machines in” filters the 4 of 60 offers whose provider says where the machine is; the others are under “Place not stated”.

60 offers; rates are per GPU and the node total is for the smallest listed node.

GPU / providerBilling$/GPU-hourNode · total per hourRent

Tables rank by price, lowest first. Commercial relationships never affect ordering. How we collect and rank prices.

Same GPU, provider, billing and price on several hosts or listings: one row, with the count. – means the provider does not state it.

Cite this data

GPU Rental Prices, "Qwen3 Coder 480B A35B VRAM: memory estimates and GPU rental costs", prices checked 22 October 2026, https://gpurentalprices.com/models/qwen3-coder-480b-a35b-instruct

Quote it with a link to this page. Downloads of the whole dataset are on Data & API.

Price a training run of Qwen3 Coder 480B A35B and check it fits the node · Compare an API bill with self-hosting

Other models from this publisher

Compare your offers

Saved prices keep their original dates. Confirm the configuration, terms and availability at the source.

Select up to three offers to compare their rates and terms.

Type a GPU, a provider, a model or a question.