# Can the RTX 4060 run lumi-4b?

> Yes. The RTX 4060 runs lumi-4b fully in VRAM at Q8_0 (about 5.9 GB of 8.0 GB), at roughly 35 tokens/s.

GeForce RTX 4060: 8 GB (8 GB usable), 272 GB/s. Estimates at an 8K context.

## Every quantisation

| Quant | Quality | ≈ GB | Fit | Est. tok/s |
|---|---|---|---|---|
| Q2_K | Heavy loss | 3.09 | fits | 115 |
| Q3_K_M | Noticeable loss | 3.71 | fits | 77 |
| Q4_K_M | Small loss | 4.17 | fits | 62 |
| Q5_K_M | Near-lossless | 4.58 | fits | 53 |
| Q6_K | Near-lossless | 5.65 | fits | 46 |
| Q8_0 (pick) | Lossless | 5.94 | fits | 35 |
| F16 | Lossless | 9.59 | offload | 12 |

## Run it

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

```bash
llama-server -hf Mateus-Nery/lumi-4b-GGUF:Q6_K
```

Serve an OpenAI-compatible API (vllm):

```bash
vllm serve ceia-br/lumi-4b
```

Smaller FP8 build (vllm):

```bash
vllm serve liodon-ai/lumi-4b-FP8
```

Docker Compose (NVIDIA) (vllm):

```yaml
services:
  vllm:
    image: vllm/vllm-openai:latest
    command: ceia-br/lumi-4b
    ports:
      - "8000:8000"
    ipc: host
    volumes:
      - ~/.cache/huggingface:/root/.cache/huggingface
    deploy:
      resources:
        reservations:
          devices:
            - driver: nvidia
              count: all
              capabilities: [gpu]
```

Serve an OpenAI-compatible API (sglang):

```bash
sglang serve ceia-br/lumi-4b --host 0.0.0.0 --port 30000
```

Docker Compose (NVIDIA) (sglang):

```yaml
services:
  sglang:
    image: lmsysorg/sglang:latest
    command: sglang serve ceia-br/lumi-4b --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 models that fit comfortably

- [Qwen/Qwen3-4B](https://selfhost.directory/models/qwen/qwen3-4b/on/rtx-4060): Q8_0 ≈5.82 GB, est. 35 tok/s
- [meta-llama/Llama-3.2-3B-Instruct](https://selfhost.directory/models/meta-llama/llama-3.2-3b-instruct/on/rtx-4060): Q8_0 ≈4.73 GB, est. 44 tok/s
- [microsoft/Phi-4-mini-instruct](https://selfhost.directory/models/microsoft/phi-4-mini-instruct/on/rtx-4060): Q8_0 ≈5.49 GB, est. 37 tok/s
- [Qwen/Qwen3-4B-Instruct-2507](https://selfhost.directory/models/qwen/qwen3-4b-instruct-2507/on/rtx-4060): Q8_0 ≈5.82 GB, est. 35 tok/s
- [XHToken/Spark-X2.6-4B](https://selfhost.directory/models/xhtoken/spark-x2.6-4b/on/rtx-4060): Q8_0 ≈5.9 GB, est. 34 tok/s
- [Qwen/Qwen2.5-3B-Instruct](https://selfhost.directory/models/qwen/qwen2.5-3b-instruct/on/rtx-4060): Q8_0 ≈4.32 GB, est. 46 tok/s

## Other GPUs that run it

- [Apple M4 (16 GB)](https://selfhost.directory/models/ceia-br/lumi-4b/on/apple-m4-16gb): F16, est. 8.3 tok/s
- [Radeon RX 6700 XT](https://selfhost.directory/models/ceia-br/lumi-4b/on/rx-6700-xt): F16, est. 27 tok/s
- [GeForce RTX 3060 12GB](https://selfhost.directory/models/ceia-br/lumi-4b/on/rtx-3060): F16, est. 25 tok/s
- [Arc B580](https://selfhost.directory/models/ceia-br/lumi-4b/on/arc-b580): F16, est. 32 tok/s
- [Tesla T4](https://selfhost.directory/models/ceia-br/lumi-4b/on/tesla-t4): F16, est. 22 tok/s
- [Arc A770 16GB](https://selfhost.directory/models/ceia-br/lumi-4b/on/arc-a770): F16, est. 39 tok/s
- [Radeon RX 7800 XT](https://selfhost.directory/models/ceia-br/lumi-4b/on/rx-7800-xt): F16, est. 43 tok/s
- [Tesla P40](https://selfhost.directory/models/ceia-br/lumi-4b/on/tesla-p40): F16, est. 24 tok/s

---
Model: https://selfhost.directory/models/ceia-br/lumi-4b. GPU: https://selfhost.directory/models/gpus/rtx-4060. Source: https://selfhost.directory/models/ceia-br/lumi-4b/on/rtx-4060