RTX A6000 vs RTX A4000
Both run on the same nodes, the same templates and the same tiers here — so the choice comes down to memory, bandwidth and price. Every figure below is from the catalogue or the vendor datasheet.
- RTX A6000
- $0.09/h48 GB · Ampere · 48 available
- RTX A4000
- $0.03/h16 GB · Ampere · 48 available
- Price gap
- 67%On spot, per GPU-hour
- Over 100 h
- $6.00Difference for the same GPU-hours
What actually separates them.
Four differences that change a decision, computed from the numbers rather than asserted.
Bigger models fit on the RTX A6000
48 GB against 16 GB. At FP16 that is about 19B parameters on one GPU, against 6B.
The RTX A4000 costs 67% less per hour
$0.03 against $0.09 on spot — $6.00 of difference over 100 GPU-hours.
Token generation is faster on the RTX A6000
768 GB/s against 448 GB/s. Inference reads the whole model from memory for every token, so bandwidth, not FLOPS, usually sets the ceiling.
Every figure we hold on both cards.
A row is marked only when both values exist and one is genuinely better. RTX A6000 leads on 4 rows, RTX A4000 on 4.
| RTX A6000 | RTX A4000 | |
|---|---|---|
| GPU memoryDecides which models fit at all | 48 GB GDDR6 | 16 GB GDDR6 |
| Memory bandwidthSets token throughput on inference | 768 GB/s | 448 GB/s |
| FP32 | 39 TFLOPS | 19 TFLOPS |
| 8-bit float (FP8)Halves the memory a model needs, where the stack supports it | Not supported | Not supported |
| Architecture | Ampere (2020) | Ampere (2021) |
| Board powerLower is cheaper to run at scale — for us, not for your bill | 300 W | 140 W |
| GPU-to-GPU linkOnly matters for multi-GPU training | NVLink | PCIe only |
| Node sizes | 1×, 2×, 4× | 1×, 2×, 4× |
| Spot price | $0.09/h | $0.03/h |
| On-demand price | $0.17/h | $0.05/h |
| Memory per dollarGPU memory per $1 of spot time, per hour | 533 GB | 533 GB |
| 30-day reclaim rateShare of spot instances reclaimed | <5% | <5% |
| Available now | 48 GPUs | 48 GPUs |
Specifications come from each vendor's datasheet — linked on the RTX A6000 page and the RTX A4000 page. Prices are ours, per GPU-hour, billed per minute.
The same models, on one GPU of each.
Weights only, with 20% of the memory kept free for activations and the KV cache. A cross means one GPU is not enough — the model page has the multi-GPU arithmetic.
| Model | Precision | RTX A6000 48 GB | RTX A4000 16 GB |
|---|---|---|---|
| Llama 3.1 8B8B parameters | FP16 · 16 GB | Fits | Needs 2× node |
| INT4 · 4 GB | Fits | Fits | |
| Qwen2.5 32B32B parameters | FP16 · 64 GB | Needs 2× node | Needs more than one node |
| INT4 · 16 GB | Fits | Needs 2× node | |
| Llama 3.3 70B70B parameters | FP16 · 140 GB | Needs 4× node | Needs more than one node |
| INT4 · 35 GB | Fits | Needs 4× node | |
| Mixtral 8x22B141B parameters | FP16 · 282 GB | Needs more than one node | Needs more than one node |
| INT4 · 71 GB | Needs 2× node | Needs more than one node |
The same 100 GPU-hours on each.
Compute only, billed per minute. Storage is $0.08 per GB-month on either card.
| RTX A6000 | RTX A4000 | |
|---|---|---|
| One hour, spot | $0.09 | $0.03 |
| One day, spot | $2.16 | $0.72 |
| 100 GPU-hours, spot | $9.00 | $3.00 |
| 100 GPU-hours, on-demand | $17.00 | $5.00 |
| A month, reserved | $65.70 | $21.90 |
Take the RTX A6000 when…
- Fine-tuning with LoRA and QLoRA — Comfortable for parameter-efficient training on 13B–34B models without sharding.
- Inference servers — Holds 30B-class models quantised, or several smaller models side by side.
- Multi-GPU jobs over NVLink — GPU-to-GPU traffic stays off the PCIe bus, which is what makes tensor and pipeline parallelism worth it.
Take the RTX A4000 when…
- Quantised inference — Small and medium models in 4-bit or 8-bit form, with the whole model resident in memory.
- Computer vision and batch jobs — Detection, classification, embeddings and transcoding at a very low hourly rate.
- Anything that can checkpoint — On the spot tier this model is 40% below its own on-demand price, with a 2-minute notice before a reclaim.
RTX A6000 or RTX A4000?
Which is cheaper, the RTX A6000 or the RTX A4000?
The RTX A4000, at $0.03 per GPU-hour on spot against $0.09 — 67% less. On-demand: $0.05 against $0.17.
Which one should I take for inference?
Take the one that holds the model with room for the KV cache, then the one with more memory bandwidth. Here that is the RTX A6000 on capacity (48 GB) and the RTX A6000 on bandwidth.
Can I run both on the same job?
Not inside one instance — a node holds GPUs of a single model. You can run two instances in parallel, on the same persistent volume in turn, or split the work: a big card for the model that needs capacity, a cheap one for the parts that do not.
Other comparisons: the full list · all 47 models
Try both for less than the price of deciding.
An hour on each costs $0.12 on spot. Nothing renews, and your disk moves between them.
Billed per minute from the moment the instance is reachable. Minimum credit $40, no subscription.