apepkuss79
commited on
Commit
•
f6d6d9b
1
Parent(s):
081e986
Update README.md
Browse files
README.md
CHANGED
@@ -25,49 +25,18 @@ tags:
|
|
25 |
|
26 |
[CompVis/stable-diffusion-v-1-4-original](https://huggingface.co/CompVis/stable-diffusion-v-1-4-original)
|
27 |
|
28 |
-
## Run with
|
29 |
|
30 |
-
|
31 |
-
|
32 |
-
|
33 |
-
<!-- - LlamaEdge version: [v0.12.2](https://github.com/LlamaEdge/LlamaEdge/releases/tag/0.12.2) and above
|
34 |
-
|
35 |
-
- Prompt template
|
36 |
-
|
37 |
-
- Prompt type: `chatml`
|
38 |
-
|
39 |
-
- Prompt string
|
40 |
-
|
41 |
-
```text
|
42 |
-
<|im_start|>system
|
43 |
-
{system_message}<|im_end|>
|
44 |
-
<|im_start|>user
|
45 |
-
{prompt}<|im_end|>
|
46 |
-
<|im_start|>assistant
|
47 |
-
```
|
48 |
-
|
49 |
-
- Context size: `4096`
|
50 |
|
51 |
- Run as LlamaEdge service
|
52 |
|
53 |
```bash
|
54 |
-
wasmedge --dir .:.
|
55 |
-
|
56 |
-
--
|
57 |
-
--ctx-size 4096 \
|
58 |
-
--model-name stablelm-2-12b-chat
|
59 |
```
|
60 |
|
61 |
-
- Run as LlamaEdge command app
|
62 |
-
|
63 |
-
```bash
|
64 |
-
wasmedge --dir .:. \
|
65 |
-
--nn-preload default:GGML:AUTO:stablelm-2-12b-chat-Q5_K_M.gguf \
|
66 |
-
llama-chat.wasm \
|
67 |
-
--prompt-template chatml \
|
68 |
-
--ctx-size 4096
|
69 |
-
``` -->
|
70 |
-
|
71 |
## Quantized GGUF Models
|
72 |
|
73 |
Using formats of different precisions will yield results of varying quality.
|
@@ -78,10 +47,10 @@ Using formats of different precisions will yield results of varying quality.
|
|
78 |
|
79 |
| Name | Quant method | Bits | Size | Use case |
|
80 |
| ---- | ---- | ---- | ---- | ----- |
|
81 |
-
| [stable-diffusion-v1-
|
82 |
-
| [stable-diffusion-v1-
|
83 |
-
| [stable-diffusion-v1-
|
84 |
-
| [stable-diffusion-v1-
|
85 |
-
| [stable-diffusion-v1-
|
86 |
-
| [stable-diffusion-v1-
|
87 |
-
| [stable-diffusion-v1-
|
|
|
25 |
|
26 |
[CompVis/stable-diffusion-v-1-4-original](https://huggingface.co/CompVis/stable-diffusion-v-1-4-original)
|
27 |
|
28 |
+
## Run with LlamaEdge-StableDiffusion
|
29 |
|
30 |
+
- Version: [v0.2.0](https://github.com/LlamaEdge/sd-api-server/releases/tag/0.2.0)
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
31 |
|
32 |
- Run as LlamaEdge service
|
33 |
|
34 |
```bash
|
35 |
+
wasmedge --dir .:. sd-api-server.wasm \
|
36 |
+
--model-name sd-v1.4 \
|
37 |
+
--model stable-diffusion-v1-4-Q8_0.gguf
|
|
|
|
|
38 |
```
|
39 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
40 |
## Quantized GGUF Models
|
41 |
|
42 |
Using formats of different precisions will yield results of varying quality.
|
|
|
47 |
|
48 |
| Name | Quant method | Bits | Size | Use case |
|
49 |
| ---- | ---- | ---- | ---- | ----- |
|
50 |
+
| [stable-diffusion-v1-4-Q4_0.gguf](https://huggingface.co/second-state/stable-diffusion-v-1-4-GGUF/blob/main/stable-diffusion-v1-4-Q4_0.gguf) | Q4_0 | 2 | 1.57 GB | |
|
51 |
+
| [stable-diffusion-v1-4-Q4_1.gguf](https://huggingface.co/second-state/stable-diffusion-v1-5-GGUF/blob/main/stable-diffusion-v1-4-Q4_1.gguf) | Q4_1 | 3 | 1.59 GB | |
|
52 |
+
| [stable-diffusion-v1-4-Q5_0.gguf](https://huggingface.co/second-state/stable-diffusion-v1-5-GGUF/blob/main/stable-diffusion-v1-4-Q5_0.gguf) | Q5_0 | 3 | 1.62 GB | |
|
53 |
+
| [stable-diffusion-v1-4-Q5_1.gguf](https://huggingface.co/second-state/stable-diffusion-v1-5-GGUF/blob/main/stable-diffusion-v1-4-Q5_1.gguf) | Q5_1 | 3 | 1.64 GB | |
|
54 |
+
| [stable-diffusion-v1-4-Q8_0.gguf](https://huggingface.co/second-state/stable-diffusion-v1-5-GGUF/blob/main/stable-diffusion-v1-4-Q8_0.gguf) | Q8_0 | 4 | 1.76 GB | |
|
55 |
+
| [stable-diffusion-v1-4-f16.gguf](https://huggingface.co/second-state/stable-diffusion-v1-5-GGUF/blob/main/stable-diffusion-v1-4-f16.gguf) | f16 | 4 | 2.13 GB | |
|
56 |
+
| [stable-diffusion-v1-4-f32.gguf](https://huggingface.co/second-state/stable-diffusion-v1-5-GGUF/blob/main/stable-diffusion-v1-4-f32.gguf) | f32 | 4 | 4.27 GB | |
|