---
inference: false
language:
  - zh
license: other
model_creator: hpcai-tech
model_link: https://huggingface.co/hpcai-tech/Colossal-LLaMA-2-7b-base
model_name: Colossal-LLaMA-2-7b-base
model_type: llama
pipeline_tag: text-generation
quantized_by: shaowenchen
tasks:
  - text2text-generation
tags:
  - gguf
  - llama
  - llama-2
  - chinese
---

## Provided files

| Name                                 | Quant method | Size   |
| ------------------------------------ | ------------ | ------ |
| colossal-llama-2-7b-base.Q2_K.gguf   | Q2_K         | 2.8 GB |
| colossal-llama-2-7b-base.Q3_K.gguf   | Q3_K         | 3.2 GB |
| colossal-llama-2-7b-base.Q3_K_L.gguf | Q3_K_L       | 3.5 GB |
| colossal-llama-2-7b-base.Q3_K_S.gguf | Q3_K_S       | 2.9 GB |
| colossal-llama-2-7b-base.Q4_0.gguf   | Q4_0         | 3.8 GB |
| colossal-llama-2-7b-base.Q4_1.gguf   | Q4_1         | 4.2 GB |
| colossal-llama-2-7b-base.Q4_K.gguf   | Q4_K         | 4.0 GB |
| colossal-llama-2-7b-base.Q4_K_S.gguf | Q4_K_S       | 3.8 GB |
| colossal-llama-2-7b-base.Q5_0.gguf   | Q5_0         | 4.5 GB |
| colossal-llama-2-7b-base.Q5_1.gguf   | Q5_1         | 4.9 GB |
| colossal-llama-2-7b-base.Q5_K.gguf   | Q5_K         | 4.7 GB |
| colossal-llama-2-7b-base.Q5_K_S.gguf | Q5_K_S       | 4.5 GB |
| colossal-llama-2-7b-base.Q6_K.gguf   | Q6_K         | 5.4 GB |
| colossal-llama-2-7b-base.Q8_0.gguf   | Q8_0         | 7.0 GB |
| colossal-llama-2-7b-base.gguf        | full         | 13 GB  |

Usage:

```bash
docker run --rm -it -p 8000:8000 -v /path/to/models:/models -e MODEL=/models/gguf-model-name.gguf shaowenchen/llama-cpp-python:0.2.6
```

## Provided images

| Name                                               | Quant method | Compressed Size |
| -------------------------------------------------- | ------------ | --------------- |
| `shaowenchen/colossal-llama-2-7b-base-gguf:Q2_K`   | Q2_K         | 3.24 GB         |
| `shaowenchen/colossal-llama-2-7b-base-gguf:Q3_K`   | Q3_K         | 3.68 GB         |
| `shaowenchen/colossal-llama-2-7b-base-gguf:Q3_K_L` | Q3_K_L       | 3.98 GB         |
| `shaowenchen/colossal-llama-2-7b-base-gguf:Q3_K_S` | Q3_K_S       | 3.38 GB         |
| `shaowenchen/colossal-llama-2-7b-base-gguf:Q4_0`   | Q4_0         | 4.05 GB         |
| `shaowenchen/colossal-llama-2-7b-base-gguf:Q4_1`   | Q4_1         | 4.47 GB         |
| `shaowenchen/colossal-llama-2-7b-base-gguf:Q4_K`   | Q4_K         | 4.39 GB         |
| `shaowenchen/colossal-llama-2-7b-base-gguf:Q4_K_S` | Q4_K_S       | 4.18 GB         |
| `shaowenchen/colossal-llama-2-7b-base-gguf:Q5_0`   | Q5_0         | 4.99 GB         |
| `shaowenchen/colossal-llama-2-7b-base-gguf:Q5_1`   | Q5_1         | 5.35 GB         |
| `shaowenchen/colossal-llama-2-7b-base-gguf:Q5_K`   | Q5_K         | 5.12 GB         |
| `shaowenchen/colossal-llama-2-7b-base-gguf:Q5_K_S` | Q5_K_S       | 5 GB            |
| `shaowenchen/colossal-llama-2-7b-base-gguf:Q6_K`   | Q6_K         | 5.82 GB         |
| `shaowenchen/colossal-llama-2-7b-base-gguf:Q8_0`   | Q8_0         | 7.18 GB         |
| `shaowenchen/colossal-llama-2-7b-base-gguf:full`   | full         | 10.49 GB        |

Usage:

```
docker run --rm -p 8000:8000 shaowenchen/colossal-llama-2-7b-base-gguf:Q2_K
```

and you can view http://localhost:8000/docs to see the swagger UI.