-
Notifications
You must be signed in to change notification settings - Fork 0
/
Copy pathdocker-compose.yml
93 lines (93 loc) · 3.94 KB
/
docker-compose.yml
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
version: "3.8"
services:
cosmos:
image: ghcr.io/ajbouh/cosmos:3.1.3
build:
dockerfile: Dockerfile
target: cosmos
python:
image: ghcr.io/ajbouh/cosmos:python-cosmo-3.1.3
build:
dockerfile: Dockerfile
target: ape
args:
COSMOS_EXE: /usr/bin/python
lua:
image: ghcr.io/ajbouh/cosmos:lua-cosmo-3.1.3
build:
dockerfile: Dockerfile
target: ape
args:
COSMOS_EXE: /usr/bin/lua
sqlite3:
image: ghcr.io/ajbouh/cosmos:sqlite3-cosmo-3.1.3
build:
dockerfile: Dockerfile
target: ape
args:
COSMOS_EXE: /usr/bin/sqlite3
qjs:
image: ghcr.io/ajbouh/cosmos:qjs-cosmo-3.1.3
build:
dockerfile: Dockerfile
target: ape
args:
COSMOS_EXE: /usr/bin/qjs
# mistral-7b-instruct-v0.1-q4_k_m-cuda:
# image: ghcr.io/ajbouh/cosmos:mistral-7b-instruct-v0.1-q4_k_m-cuda-12.1.1-cosmo-3.1.3
# deploy: {resources: {reservations: {devices: [{driver: nvidia, count: all, capabilities: ["gpu"]}]}}}
# build:
# dockerfile: Dockerfile
# target: llamafile-cuda
# args:
# LLAMAFILE_URL: https://huggingface.co/jartine/mistral-7b.llamafile/resolve/649327b402e83a9d251adf813bc5f64fee5dbdd3/mistral-7b-instruct-v0.1-Q4_K_M-main.llamafile?download=true
# LLAMAFILE_CHECKSUM: sha256:1944286a05b979cb37652c2cf8a00f3fbc5275b6d0108b36adb199962de65562
# LLAMAFILE_N_GPU_LAYERS: 35
# mistral-7b-instruct-v0.1-q4_k_m:
# image: ghcr.io/ajbouh/cosmos:mistral-7b-instruct-v0.1-q4_k_m-cosmo-3.1.3
# build:
# dockerfile: Dockerfile
# target: llamafile
# args:
# LLAMAFILE_URL: https://huggingface.co/jartine/mistral-7b.llamafile/resolve/649327b402e83a9d251adf813bc5f64fee5dbdd3/mistral-7b-instruct-v0.1-Q4_K_M-main.llamafile?download=true
# LLAMAFILE_CHECKSUM: sha256:1944286a05b979cb37652c2cf8a00f3fbc5275b6d0108b36adb199962de65562
llava-v1.5-7b-q4_k-cuda:
image: ghcr.io/ajbouh/cosmos:llava-v1.5-7b-q4_k-cuda-12.1.1-cosmo-3.1.3
deploy: {resources: {reservations: {devices: [{driver: nvidia, count: all, capabilities: ["gpu"]}]}}}
ports: ["8080:8080"]
build:
dockerfile: Dockerfile
target: llamafile-gguf-cuda
args:
GGUF_URL: https://huggingface.co/jartine/llava-v1.5-7B-GGUF/resolve/main/llava-v1.5-7b-Q4_K.gguf?download=true
GGUF_CHECKSUM: sha256:c91ebf0a628ceb25e374df23ad966cc1bf1514b33fecf4f0073f9619dec5b3f9
LLAMAFILE_N_GPU_LAYERS: 35
llava-v1.5-7b-q4_k:
image: ghcr.io/ajbouh/cosmos:llava-v1.5-7b-q4_k-cosmo-3.1.3
ports: ["8080:8080"]
build:
dockerfile: Dockerfile
target: llamafile-gguf
args:
GGUF_URL: https://huggingface.co/jartine/llava-v1.5-7B-GGUF/resolve/main/llava-v1.5-7b-Q4_K.gguf?download=true
GGUF_CHECKSUM: sha256:c91ebf0a628ceb25e374df23ad966cc1bf1514b33fecf4f0073f9619dec5b3f9
airoboros-m-7b-3.1.2-dare-0.85.q4_k_m-cuda:
image: ghcr.io/ajbouh/cosmos:airoboros-m-7b-3.1.2-dare-0.85.q4_k_m-cuda-12.1.1-cosmo-3.1.3
deploy: {resources: {reservations: {devices: [{driver: nvidia, count: all, capabilities: ["gpu"]}]}}}
ports: ["8080:8080"]
build:
dockerfile: Dockerfile
target: llamafile-gguf-cuda
args:
GGUF_URL: https://huggingface.co/TheBloke/airoboros-m-7B-3.1.2-dare-0.85-GGUF/resolve/main/airoboros-m-7b-3.1.2-dare-0.85.Q4_K_M.gguf?download=true
GGUF_CHECKSUM: sha256:5d6bc74b99aa89d3c35c90c74d6844e1e45bd810dd08f9f55252f74ed87b0663
LLAMAFILE_N_GPU_LAYERS: 35
airoboros-m-7b-3.1.2-dare-0.85.q4_k_m:
image: ghcr.io/ajbouh/cosmos:airoboros-m-7b-3.1.2-dare-0.85.q4_k_m-cosmo-3.1.3
ports: ["8080:8080"]
build:
dockerfile: Dockerfile
target: llamafile-gguf
args:
GGUF_URL: https://huggingface.co/TheBloke/airoboros-m-7B-3.1.2-dare-0.85-GGUF/resolve/main/airoboros-m-7b-3.1.2-dare-0.85.Q4_K_M.gguf?download=true
GGUF_CHECKSUM: sha256:5d6bc74b99aa89d3c35c90c74d6844e1e45bd810dd08f9f55252f74ed87b0663