RTX 6000 Ada Specifications
48 GB ECC with FP8 at 300 W. The strongest workstation card for local training before you move to datacenter parts.
Buy it for
Local AI workstations needing 48 GB with ECC, in a machine that plugs into a normal wall socket.
Look elsewhere if
Budget is the constraint — two used A100s cost less and deliver far more for training.
Will it fit your system?
- Form factor
- PCIe dual-slot
- Cooling
- active
- Power connectors
- 1x 16-pin CEM5 (12VHPWR)
- Board power
- 300 W
- Suggested PSU
- 750 W or better
- Interface
- PCIe Gen 4
PSU figure assumes one card plus a mainstream CPU and drives, with headroom for transient spikes — modern cards can draw well above their rated TDP for milliseconds, which is what trips an undersized supply.
Supported formats & features
What the RTX 6000 Ada does and does not support in hardware.
Blackwell-generation feature — B200 has it
E4M3 and E5M2 — the standard format for modern inference
Wider exponent than FP16 — fewer overflow issues when training
Drop-in FP32 speedup with no code changes
Quantised inference supported
Not yet confirmed against vendor documentation.
Precision must be managed manually
Multi-GPU traffic goes over PCIe — a real limit for distributed training
Hardware ray tracing for rendering workloads
Corrects single-bit errors — expected for long training runs
What fits in memory
Largest model that fits on one RTX 6000 Ada, assuming ~20% of the 48 GB is reserved for KV cache and activations.
| Model | FP16 / BF16 | FP8 |
|---|---|---|
| Llama 8B | Fits | Fits |
| Llama 70B | 4× GPUs | 2× GPUs |
| Mixtral 8x22B | 8× GPUs | 4× GPUs |
| Llama 405B | 22× GPUs | 11× GPUs |
Weights only. Long contexts, large batch sizes and optimiser state during training all need substantially more — full fine-tuning typically needs 3–4× the weight memory.
Workload character
Plenty of bandwidth per unit of compute. Memory-bound work — single-stream token decode above all — runs close to its ceiling, and compute is usually what runs out first.
Cost to run
Electricity for one card at 100% duty, including cooling and PSU overhead (3,679 kWh/year).
At the US average rate, three years of continuous operation costs $1,325 — roughly 19% of the card's $6,800 launch price, before any cooling or hosting fees.
Availability & market
Sold through workstation OEMs and channel resellers. Holds value well.
Memory
- Capacity
- 48 GB
- Type
- GDDR6 ECC
- Bandwidth
- 960 GB/s
- Bus width
- 384-bit
- GB per TFLOPMemory relative to compute — higher favours large models
- 0.264
Compute
- FP64
- —
- FP32
- 91.1 TFLOPS
- TF32 dense
- 182 TFLOPS
- FP16 dense
- 182 TFLOPS
- FP16 sparseRequires structured sparsity; rarely reached in practice
- 364 TFLOPS
- FP8 dense
- 728 TFLOPS
- INT8 dense
- 728 TOPS
Silicon
- Codename
- AD102
- Process node
- TSMC 4N
- Transistors
- 76.3B
- Die size
- 609 mm²
- CUDA cores
- 18,176
- Tensor cores
- 568
- SM count
- 142
- Boost clock
- 2,505 MHz
Platform & power
- TDP
- 300 W
- TFLOPS per wattDense FP16 ÷ TDP — drives your power bill at scale
- 0.61
- Form factor
- PCIe dual-slot
- Cooling
- active
- Interconnect
- None
- Interconnect speed
- —
- PCIe generation
- Gen 4
- MIG
- No
- Launch
- 2022-12
- MSRP at launch
- $6,800
RTX 6000 Ada — common questions
Answered from the specifications above. Questions this card's data cannot answer with confidence are left out rather than guessed.
- How much VRAM does the RTX 6000 Ada have?
- The RTX 6000 Ada has 48 GB of GDDR6 ECC memory with 960 GB/s of bandwidth. In practice plan on roughly 38 GB for weights: KV cache, activations and fragmentation take the rest, and long contexts or large batches take considerably more.
- What size model can the RTX 6000 Ada run?
- On a single RTX 6000 Ada, about 19B parameters at FP16, or roughly 38B quantised to FP8. Anything larger has to be split across several GPUs, which is slow on this card because it has no NVLink. These are weight-only figures — a long context window can consume tens of gigabytes on its own.
- Can you put the RTX 6000 Ada in a desktop PC?
- Yes. The RTX 6000 Ada is a self-cooled PCIe dual-slot card and works in a normal desktop, provided the case has clearance and the power supply is rated around 750 W or better.
- How much power does the RTX 6000 Ada use?
- The RTX 6000 Ada is rated at 300 W board power. It is fed through 1x 16-pin CEM5 (12VHPWR). For a single-card system, a 750 W supply or better leaves headroom for the CPU, drives and the transient spikes that trip undersized units.
- How much does it cost to run the RTX 6000 Ada?
- At the US commercial average of $0.12/kWh, running one RTX 6000 Ada continuously costs about $442 a year in electricity. That figure includes cooling and power-supply overhead at a factor of 1.4 over board power, and over three years comes to roughly 19% of what the card cost at launch.
- Does the RTX 6000 Ada support FP8?
- Yes. The RTX 6000 Ada has hardware FP8 tensor support, so FP8 weights and activations run at full tensor-core rate — 728 dense TFLOPS. This is the format most current inference servers default to for large models.
- Does the RTX 6000 Ada support FP4?
- No. The RTX 6000 Ada predates hardware FP4. You can run 4-bit quantised weights on it and the memory saving is real, but they are dequantised before compute, so there is no arithmetic speedup.
- Does the RTX 6000 Ada support INT8?
- Yes. The RTX 6000 Ada supports INT8 acceleration at 728 dense TOPS. INT8 remains the most widely supported quantised inference format across frameworks.
- Does the RTX 6000 Ada support BF16?
- Yes. The RTX 6000 Ada supports bfloat16. Its wider exponent range makes training far less prone to the overflow that plagues FP16, which is why most current training recipes assume it.
- Does the RTX 6000 Ada have NVLink?
- No. The RTX 6000 Ada has no NVLink. Multi-GPU work goes over PCIe Gen 4, which is an order of magnitude slower and makes tensor parallelism across cards expensive.
- Does the RTX 6000 Ada have RT cores?
- Yes. The RTX 6000 Ada includes dedicated ray-tracing cores, so it is usable for rendering and visualisation as well as compute.
- Does the RTX 6000 Ada have ECC memory?
- Yes. The RTX 6000 Ada has ECC, so single-bit memory errors are corrected rather than silently corrupting a long run.
- Is the RTX 6000 Ada worth buying?
- Buy it for: Local AI workstations needing 48 GB with ECC, in a machine that plugs into a normal wall socket. Look elsewhere if: Budget is the constraint — two used A100s cost less and deliver far more for training.