# NVIDIA RTX 6000 Ada Generation for local LLMs

> 48GB in a 300W blower card that stacks well, though newer options undercut it.

The RTX 6000 Ada has 48GB of GDDR6 ECC and 960 GB/s of memory bandwidth, which decides how big a model fits and how fast it generates text. It comfortably runs models up to about 35B parameters: the largest in our catalog is Qwen3.6 35B-A3B at Q8_0 (41GB with 8K context, ~100 tokens/sec). Our top all-round pick for it is Qwen3.8 27B (Q8_0, ~25 tokens/sec).

## Key specs

- Memory: 48GB GDDR6 ECC
- Memory bandwidth: 960 GB/s
- Architecture: Ada Lovelace (2022)
- Power: 300W
- Typical price: ~$4,750 used (checked 2026-10)
- Software: CUDA, Vulkan

## Best models for this GPU

- **Best overall (also coding, reasoning, vision):** Qwen3.8 27B at Q8_0, ~31.6GB with 8K context, ~25 tokens/sec
- **Runner-up (also largest that fits):** Qwen3.6 35B-A3B at Q8_0, ~41.0GB with 8K context, ~100 tokens/sec
- **Fastest good model:** gpt-oss-20b at MXFP4, ~15.0GB with 8K context, ~130 tokens/sec

## Compatibility (best quantization that fits, 8K context)

| Model | Params | Quant | Memory | Speed | Verdict |
| --- | --- | --- | --- | --- | --- |
| Gemma 4 E4B | 8B | Q8_0 | 10.5GB | ~75 tok/s | Runs well |
| Qwen3.5 9B | 9B | Q8_0 | 12.0GB | ~66 tok/s | Runs well |
| Phi-4 | 14.7B | Q8_0 | 18.3GB | ~47 tok/s | Runs well |
| gpt-oss-20b | 20.9B (3.6B active) | MXFP4 | 15.0GB | ~130 tok/s | Runs well |
| Gemma 4 26B A4B | 26B (3.8B active) | Q8_0 | 29.5GB | ~89 tok/s | Runs well |
| Qwen3.8 27B | 27B | Q8_0 | 31.6GB | ~25 tok/s | Runs well |
| Gemma 4 31B | 31B | Q8_0 | 35.9GB | ~23 tok/s | Runs well |
| Qwen3.6 35B-A3B | 35B (3B active) | Q8_0 | 41.0GB | ~100 tok/s | Runs well |
| GLM-4.5-Air | 106B (12B active) | Q4_K_M | 68.3GB | ~15 tok/s | Partial offload |
| Llama 4 Scout | 109B (17B active) | Q4_K_M | 69.5GB | ~11 tok/s | Partial offload |
| gpt-oss-120b | 117B (5.1B active) | MXFP4 | 68.3GB | ~37 tok/s | Partial offload |
| Qwen3.5 122B-A10B | 122B (10B active) | Q4_K_M | 77.0GB | ~16 tok/s | Partial offload |
| DeepSeek V4 Flash | 284B (13B active) | - | - | - | Won't fit |
| GLM-5.3-Flash | 320B (18B active) | - | - | - | Won't fit |
| GLM-5.3 | 744B (40B active) | - | - | - | Won't fit |

## Pros
- 48GB per card with ECC
- 300W blower design stacks densely
- 960 GB/s bandwidth

## Cons
- Used prices remain high
- Slower than Blackwell workstation cards
- No NVLink

Full page: https://llamabuilds.ai/gpus/rtx-6000-ada · Best models: https://llamabuilds.ai/gpus/rtx-6000-ada/best-models

Sources: [NVIDIA RTX 6000 Ada Generation product page](https://www.nvidia.com/en-us/products/workstations/rtx-6000/)

---
Source: https://llamabuilds.ai/gpus/rtx-6000-ada · Updated October 14, 2026 · Llama Builds. Speeds marked ~ are estimates from our bandwidth model (https://llamabuilds.ai/methodology); measured results are labeled as such.