# Can the RTX 4070 run Spark-X2.6-1.7B?

> Yes. The RTX 4070 runs Spark-X2.6-1.7B fully in VRAM at FP16 / BF16 (about 4.3 GB of 12 GB), at roughly 81 tokens/s.

GeForce RTX 4070: 12 GB (12 GB usable), 504 GB/s. Estimates at an 8K context.

## Every quantisation

| Quant | Quality | ≈ GB | Fit | Est. tok/s |
|---|---|---|---|---|
| Q2_K | Heavy loss | 1.49 | fits | 494 |
| Q3_K_M | Noticeable loss | 1.75 | fits | 332 |
| Q4_K_M | Small loss | 2.67 | fits | 268 |
| Q5_K_M | Near-lossless | 2.13 | fits | 228 |
| Q6_K | Near-lossless | 2.31 | fits | 197 |
| Q8_0 | Lossless | 2.73 | fits | 153 |
| F16 (pick) | Lossless | 4.28 | fits | 81 |

## Run it

Q8_0 straight from Hugging Face (ollama):

```bash
ollama run hf.co/XHToken/Spark-X2.6-1.7B-GGUF:Q8_0
```

OpenAI-compatible server with a web UI (llama.cpp):

```bash
llama-server -hf XHToken/Spark-X2.6-1.7B-GGUF:Q8_0
```

Download with the lms CLI (lmstudio):

```bash
lms get XHToken/Spark-X2.6-1.7B-GGUF@Q8_0
```

Serve an OpenAI-compatible API (sglang):

```bash
sglang serve XHToken/Spark-X2.6-1.7B --host 0.0.0.0 --port 30000
```

Docker Compose (NVIDIA) (sglang):

```yaml
services:
  sglang:
    image: lmsysorg/sglang:latest
    command: sglang serve XHToken/Spark-X2.6-1.7B --host 0.0.0.0 --port 30000
    ports:
      - "30000:30000"
    ipc: host
    volumes:
      - ~/.cache/huggingface:/root/.cache/huggingface
    deploy:
      resources:
        reservations:
          devices:
            - driver: nvidia
              count: all
              capabilities: [gpu]
```

Chat with the pipeline API (transformers):

```python
from transformers import pipeline

pipe = pipeline("text-generation", model="XHToken/Spark-X2.6-1.7B", 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

- [LiquidAI/LFM3-1.7B](https://selfhost.directory/models/liquidai/lfm3-1.7b/on/rtx-4070): Q8_0 ≈2.69 GB, est. 151 tok/s
- [Qwen/Qwen2.5-1.5B-Instruct](https://selfhost.directory/models/qwen/qwen2.5-1.5b-instruct/on/rtx-4070): Q8_0 ≈2.57 GB, est. 169 tok/s
- [Qwen/Qwen3-1.7B](https://selfhost.directory/models/qwen/qwen3-1.7b/on/rtx-4070): Q8_0 ≈3.18 GB, est. 128 tok/s
- [meta-llama/Llama-3.2-1B-Instruct](https://selfhost.directory/models/meta-llama/llama-3.2-1b-instruct/on/rtx-4070): Q8_0 ≈2.04 GB, est. 211 tok/s
- [google/gemma-3-1b-it](https://selfhost.directory/models/google/gemma-3-1b-it/on/rtx-4070): Q8_0 ≈1.75 GB, est. 261 tok/s
- [openbmb/MiniCPM5-2B](https://selfhost.directory/models/openbmb/minicpm5-2b/on/rtx-4070): Q8_0 ≈3.46 GB, est. 104 tok/s

## Other GPUs that run it

- [GeForce RTX 4060](https://selfhost.directory/models/xhtoken/spark-x2.6-1.7b/on/rtx-4060): F16, est. 44 tok/s
- [Apple M4 (16 GB)](https://selfhost.directory/models/xhtoken/spark-x2.6-1.7b/on/apple-m4-16gb): F16, est. 19 tok/s
- [Radeon RX 6700 XT](https://selfhost.directory/models/xhtoken/spark-x2.6-1.7b/on/rx-6700-xt): F16, est. 62 tok/s
- [GeForce RTX 3060 12GB](https://selfhost.directory/models/xhtoken/spark-x2.6-1.7b/on/rtx-3060): F16, est. 58 tok/s
- [Arc B580](https://selfhost.directory/models/xhtoken/spark-x2.6-1.7b/on/arc-b580): F16, est. 73 tok/s
- [Tesla T4](https://selfhost.directory/models/xhtoken/spark-x2.6-1.7b/on/tesla-t4): F16, est. 52 tok/s
- [Arc A770 16GB](https://selfhost.directory/models/xhtoken/spark-x2.6-1.7b/on/arc-a770): F16, est. 90 tok/s
- [Radeon RX 7800 XT](https://selfhost.directory/models/xhtoken/spark-x2.6-1.7b/on/rx-7800-xt): F16, est. 100 tok/s

---
Model: https://selfhost.directory/models/xhtoken/spark-x2.6-1.7b. GPU: https://selfhost.directory/models/gpus/rtx-4070. Source: https://selfhost.directory/models/xhtoken/spark-x2.6-1.7b/on/rtx-4070