# Can the RTX A4000 run RWKV7-G1k-13.3B-20260930?

> Yes, but only at Q3_K_M (about 14 GB of 16 GB), which costs noticeable quality, at roughly 38 tokens/s.

RTX A4000: 16 GB (16 GB usable), 448 GB/s. Estimates at an 8K context.

## Every quantisation

| Quant | Quality | ≈ GB | Fit | Est. tok/s |
|---|---|---|---|---|
| Q2_K | Heavy loss | 12.39 | fits | 56 |
| Q3_K_M (pick) | Noticeable loss | 14.47 | fits | 38 |
| Q4_K_M | Small loss | 15.99 | tight | 31 |
| Q5_K_M | Near-lossless | 17.35 | offload | 17 |
| Q6_K | Near-lossless | 18.81 | offload | 11 |
| Q8_0 | Lossless | 21.91 | offload | 6.4 |
| F16 | Lossless | 34.08 | offload | 2.1 |

## Run it

Chat with the pipeline API (transformers):

```python
from transformers import pipeline

pipe = pipeline("text-generation", model="RWKV/RWKV7-G1k-13.3B-20260930", dtype="auto", device_map="auto")
messages = [{"role": "user", "content": "Explain self-hosting in one paragraph."}]
out = pipe(messages, max_new_tokens=512)
print(out[0]["generated_text"][-1]["content"])
```


## Chat models that fit comfortably

- [Qwen/Qwen3-14B](https://selfhost.directory/models/qwen/qwen3-14b/on/rtx-a4000): Q6_K ≈13.6 GB, est. 20 tok/s
- [microsoft/phi-4](https://selfhost.directory/models/microsoft/phi-4/on/rtx-a4000): Q6_K ≈13.82 GB, est. 20 tok/s
- [openai/gpt-oss-20b](https://selfhost.directory/models/openai/gpt-oss-20b/on/rtx-a4000): Q8_0 ≈12.72 GB, est. 55 tok/s
- [Qwen/Qwen3-8B](https://selfhost.directory/models/qwen/qwen3-8b/on/rtx-a4000): Q8_0 ≈10.15 GB, est. 28 tok/s
- [meta-llama/Llama-3.1-8B-Instruct](https://selfhost.directory/models/meta-llama/llama-3.1-8b-instruct/on/rtx-a4000): Q8_0 ≈9.84 GB, est. 29 tok/s
- [Qwen/Qwen2.5-14B-Instruct](https://selfhost.directory/models/qwen/qwen2.5-14b-instruct/on/rtx-a4000): Q6_K ≈13.85 GB, est. 20 tok/s

## Other GPUs that run it

- [Tesla T4](https://selfhost.directory/models/rwkv/rwkv7-g1k-13.3b-20260930/on/tesla-t4): Q3_K_M, est. 27 tok/s
- [Arc A770 16GB](https://selfhost.directory/models/rwkv/rwkv7-g1k-13.3b-20260930/on/arc-a770): Q3_K_M, est. 47 tok/s
- [Radeon RX 7800 XT](https://selfhost.directory/models/rwkv/rwkv7-g1k-13.3b-20260930/on/rx-7800-xt): Q3_K_M, est. 53 tok/s
- [Tesla P40](https://selfhost.directory/models/rwkv/rwkv7-g1k-13.3b-20260930/on/tesla-p40): Q8_0, est. 14 tok/s
- [Radeon RX 7900 XTX](https://selfhost.directory/models/rwkv/rwkv7-g1k-13.3b-20260930/on/rx-7900-xtx): Q8_0, est. 37 tok/s
- [Instinct MI50 32GB](https://selfhost.directory/models/rwkv/rwkv7-g1k-13.3b-20260930/on/instinct-mi50): Q8_0, est. 40 tok/s
- [GeForce RTX 5090](https://selfhost.directory/models/rwkv/rwkv7-g1k-13.3b-20260930/on/rtx-5090): Q8_0, est. 70 tok/s
- [Apple M4 Pro (48 GB)](https://selfhost.directory/models/rwkv/rwkv7-g1k-13.3b-20260930/on/apple-m4-pro-48gb): F16, est. 5.7 tok/s

---
Model: https://selfhost.directory/models/rwkv/rwkv7-g1k-13.3b-20260930. GPU: https://selfhost.directory/models/gpus/rtx-a4000. Source: https://selfhost.directory/models/rwkv/rwkv7-g1k-13.3b-20260930/on/rtx-a4000