# NVIDIA RTX A6000 for local LLMs

> A proven 48GB Ampere workstation card with NVLink, now priced near its launch MSRP.

The RTX A6000 has 48GB of GDDR6 ECC and 768 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, ~79 tokens/sec). Our top all-round pick for it is Qwen3.8 27B (Q8_0, ~18 tokens/sec).

## Key specs

- Memory: 48GB GDDR6 ECC
- Memory bandwidth: 768 GB/s
- Architecture: Ampere (2020)
- Power: 300W
- Typical price: ~$4,400 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, ~18 tokens/sec
- **Runner-up (also largest that fits):** Qwen3.6 35B-A3B at Q8_0, ~41.0GB with 8K context, ~79 tokens/sec
- **Fastest good model:** gpt-oss-20b at MXFP4, ~15.0GB with 8K context, ~105 tokens/sec

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

| Model | Params | Quant | Memory | Speed | Verdict |
| --- | --- | --- | --- | --- | --- |
| Gemma 4 E4B | 8B | Q8_0 | 10.5GB | ~55 tok/s | Runs well |
| Qwen3.5 9B | 9B | Q8_0 | 12.0GB | ~49 tok/s | Runs well |
| Phi-4 | 14.7B | Q8_0 | 18.3GB | ~34 tok/s | Runs well |
| gpt-oss-20b | 20.9B (3.6B active) | MXFP4 | 15.0GB | ~105 tok/s | Runs well |
| Gemma 4 26B A4B | 26B (3.8B active) | Q8_0 | 29.5GB | ~68 tok/s | Runs well |
| Qwen3.8 27B | 27B | Q8_0 | 31.6GB | ~18 tok/s | Runs well |
| Gemma 4 31B | 31B | Q8_0 | 35.9GB | ~17 tok/s | Runs well |
| Qwen3.6 35B-A3B | 35B (3B active) | Q8_0 | 41.0GB | ~79 tok/s | Runs well |
| GLM-4.5-Air | 106B (12B active) | Q4_K_M | 68.3GB | ~14 tok/s | Partial offload |
| Llama 4 Scout | 109B (17B active) | Q4_K_M | 69.5GB | ~10 tok/s | Partial offload |
| gpt-oss-120b | 117B (5.1B active) | MXFP4 | 68.3GB | ~35 tok/s | Partial offload |
| Qwen3.5 122B-A10B | 122B (10B active) | Q4_K_M | 77.0GB | ~15 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
- Two-way NVLink support
- 300W blower design for multi-GPU towers

## Cons
- 768 GB/s is slower than GeForce cards with less memory
- Used prices close to launch MSRP

Full page: https://llamabuilds.ai/gpus/rtx-a6000 · Best models: https://llamabuilds.ai/gpus/rtx-a6000/best-models

Sources: [NVIDIA RTX A6000 product page](https://www.nvidia.com/en-us/design-visualization/rtx-a6000/)

---
Source: https://llamabuilds.ai/gpus/rtx-a6000 · Updated October 14, 2026 · Llama Builds. Speeds marked ~ are estimates from our bandwidth model (https://llamabuilds.ai/methodology); measured results are labeled as such.