Cheapest540 GB
2× B300
$15.78per hour
- Speed
- ~476 tok/sFaster than you read
- Per million tokens
- $9.21
RunPod · Secure Cloud
Rent on RunPodLowest price per hour
Can the AMD Radeon RX 9060 XT 16GB run DeepSeek-V3?
DeepSeek-V3 · AMD Radeon RX 9060 XT 16GB · 8,192 ctx · whole model residentQ2_K sizesize range
IT DOES NOT FIT
No. Even Q2_K, the smallest format evaluated, needs 272.5 GB against 16 GB — short by 256.5 GB. No published file exists for this pair, so every size is rebuilt from the pinned architecture.
0/9formats that fit
16 GBdevice memory
272.5 GBneeded at smallest format
—speed unavailable: does not fit
320GB/s bandwidth
Run it in the cloud
Run DeepSeek-V3 anyway, on a rented GPU
The AMD Radeon RX 9060 XT 16GB cannot hold it, but these rented machines hold the whole model. The ones worth choosing between, priced live and billed by the second.
Your card Offloading would put about 407 GB in system RAM — more than 64 GB, and at system-memory speed. For this model, a machine that holds it all on the GPU is the practical way to run it. Check with your RAM →
Prices · 18:30 UTC, 22 Sept
Recommended · Fastest720 GB
4× B200
$25.00per hour
- Speed
- ~904 tok/sFaster than you read
- Per million tokens
- $7.68
Vast.ai · Marketplace · 99.6% reliable
Rent on Vast.aiLowest cost per token at 20+ tok/s
Compare all 6 rentable configurations →
No GPU at all
Use DeepSeek-V3 by the token
- OpenRouter$0.89 / M output · $0.32 / M inputOpen on OpenRouter
For one person chatting, that is about 8.6× cheaper than the best-value rented card above ($7.68 per million tokens). A rented GPU pays off when you keep it busy — many requests batched together, long agent runs — or when the data must stay on a machine you control, or the exact file you want is not served anywhere.
First time renting a GPU? How it works, in four steps
- Create an account and add credit. Both providers are prepaid: $10 is enough to try any card on this page for hours. A new RunPod account that arrives through a referral link gets a one-time $5 credit when it first adds $10.
- Pick the card and the count shown here, and a template: llama.cpp or vLLM for an OpenAI-compatible API, or one with a web chat if you only want to talk to the model. “How to launch it” below gives the exact image and command.
- Wait for the download. The weights are fetched on the machine; tens of gigabytes take a few minutes on a data-centre connection. You pay by the second from the moment it starts.
- Stop it when you are done. A running machine bills even when idle. A stopped one costs nothing per hour, though its disk is usually still billed until you delete it.
How to launch it
No Q4_K_M file of DeepSeek-V3 is catalogued here, so there is no exact command to vouch for. Start the machine from the llama.cpp server image ghcr.io/ggml-org/llama.cpp:server-cuda and point -hf at a Q4_K_M build from the publisher or a quantizer you trust — search Hugging Face for one. Check its file size against the memory figure on this page before you rent.
Every machine holds the whole model at Q4_K_M and 8,192 tokens of context, no offload. Speeds are this site’s single-stream decode estimates; prices are what each provider’s own API quoted, on-demand, for the whole machine. Vast hosts below 98% measured reliability are left out.
Referral links Vast.ai, RunPod and Novita pay us a share of what you spend if you sign up through these buttons. It costs you nothing, and it never decides an order or a recommendation: both are computed from the live price and the speed, and options that pay us nothing are listed and recommended on the same terms. How we rank
Runs well on your AMD Radeon RX 9060 XT 16GB instead
The models nearest to DeepSeek-V3 — same lab first, then closest in size — that the AMD Radeon RX 9060 XT 16GB holds whole at the standard format and runs at a usable speed. Nearest in size is not the same as equally good; compare them on the task you care about.
- DeepSeek-R1-Distill-Qwen-14BQ4_K_M · 11.4 GB · ~25 tok/s
- DeepSeek-Coder-V2-Lite-InstructQ4_K_M · 10.7 GB · ~165 tok/s
- DeepSeek-V2-LiteQ4_K_M · 10.7 GB · ~165 tok/s
- DeepSeek-V2-Lite-ChatQ4_K_M · 10.7 GB · ~165 tok/s
Every format evaluated
No published file exists for this pair, so every size is rebuilt from the pinned architecture. The Basis column says which is which for each row. Decode and first-token figures are calibrated estimates, not runs on this card.
| Format | Needs | Of 16 GB | Fits | Decode | First token | Basis |
|---|---|---|---|---|---|---|
| FP16 | 1371.0 GB | 8569% | short by 1355.0 GB | does not run | — | size range |
| Q8_0 | 729.3 GB | 4558% | short by 713.3 GB | does not run | — | size range |
| Q6_K | 563.5 GB | 3522% | short by 547.5 GB | does not run | — | size range |
| Q5_K_M | 490.4 GB | 3065% | short by 474.4 GB | does not run | — | size range |
| Q5_0 | 479.1 GB | 2994% | short by 463.1 GB | does not run | — | size range |
| Q4_K_M | 421.6 GB | 2635% | short by 405.6 GB | does not run | — | size range |
| Q4_0 | 399.7 GB | 2498% | short by 383.7 GB | does not run | — | size range |
| Q3_K_M | 343.7 GB | 2148% | short by 327.7 GB | does not run | — | size range |
| Q2_K | 272.5 GB | 1703% | short by 256.5 GB | does not run | — | size range |
Sized at 8,192 tokens of context with the whole model resident — weights, the KV cache and the runtime reserve, offload off. Speed is only quoted for a format that fits: a rate for a configuration that cannot load is not a fact about anything.
Why the first-token figure is the same on every row
First token is modelled from the arithmetic the prompt requires, and that count does not change with the weight format — which is why it reads the same on every row. Real prefill does vary by format, because a quantized matmul is a different kernel; this model does not capture that, and the figure should be read as an order of magnitude rather than a ranking between formats.
What the labels mean
- published data
- Read from a published source — a file's byte count, a model's configuration or a manufacturer's specification — or exact arithmetic on such values. Not a measurement on a machine.
- reconstructed size
- Weight size reconstructed from the pinned architecture, because no published file exists.
- size range
- Only a lower and an upper bound are claimed for this weight size.
- estimate
- Calculated from sourced inputs by a stated method; an estimate, not a measurement.
Ways out
OPTION 1larger device
AMD Instinct MI355X
288 GB holds it at Q2_K.
OPTION 2larger device
AMD Instinct MI350X
288 GB holds it at Q2_K.
OPTION 3larger device
NVIDIA Rubin
288 GB holds it at Q2_K.
OPTION 4system RAM
Offload to system memory
Keeping some layers on the host lets it load at all, at a large cost in speed. The calculator sizes the split and prices it.
What this device does run
- Qwen3.5-9B 7.05 GB
- Qwen3.5-4B 3.98 GB
- gemma-4-12B-it 9.54 GB
- Qwen3.5-2B 2.32 GB
- gemma-4-E4B-it 5.86 GB
- NVIDIA-Nemotron-3-Nano-4B-BF16 3.46 GB
The full list is on the AMD Radeon RX 9060 XT 16GB page.
Change anything
This page fixes the context at 8,192 tokens and one device. Batch, concurrent users, KV-cache format, clusters and rental cost are all in the calculator, already set to this pairing.
Open in the calculator →Everything the AMD Radeon RX 9060 XT 16GB runsBest models for 16 GB
Where these numbers come from
DeepSeek-V3’s architecture is read from its publisher’s own config.json at a pinned revision, and the AMD Radeon RX 9060 XT 16GB’s 16 GB and 320 GB/s come from the manufacturer’s specification. How far each figure can be trusted is published, per format and worst case included, on the accuracy scorecard.
Building this into your own product? A free API key returns exactly these figures, and the widget puts this answer on a product page with one script tag.
Use this answer in your own product
As JSON, from the API
curl -s https://llmbottleneck.com/v1/analyze \
-H "Authorization: Bearer $LLMB_KEY" -H "content-type: application/json" \
-d '{"model":"deepseek-ai-deepseek-v3","quantization":"Q2_K","context":8192,"hardware":"amd-radeon-rx-9060-xt-16gb"}'Same engine, same evidence, every field sourced. Free key in one step →
On your page, as a widget
<script src="https://llmbottleneck.com/widget.js" data-model="deepseek-ai-deepseek-v3" data-quantization="Q2_K" data-hardware="amd-radeon-rx-9060-xt-16gb" data-context="8192"></script>
No key needed. Unbranded, with your own buy button, on Pro and Business →