Skip to content

Commit 5e10a46

Browse files
committed
dria: add DKN compute node
1 parent 2b2feb7 commit 5e10a46

File tree

3 files changed

+107
-0
lines changed

3 files changed

+107
-0
lines changed

README.md

Lines changed: 1 addition & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -56,6 +56,7 @@ Also, follow [@akashnet\_](https://twitter.com/akashnet_) to stay in the loop wi
5656
- [ChatChat](chatchat)
5757
- [ChatGPT Self-Hosted Chat](ai-chat-app)
5858
- [Daila](daila)
59+
- [Dria](dria)
5960
- [Flowise](flowise)
6061
- [GPT4ALL](gpt4all)
6162
- [Serge](serge-cpu)

dria/README.md

Lines changed: 11 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,11 @@
1+
# Dria Compute Node
2+
3+
A [Dria](https://dria.co/) Compute Node is a unit of computation within the Dria Knowledge Network. It's purpose is to process tasks given by the Dria Admin Node, and receive rewards for providing correct results.
4+
5+
## Setup
6+
7+
Check [dkn-compute-node](https://hub.docker.com/r/firstbatch/dkn-compute-node/tags) Docker repository to see if there is a new version and update `dkn` service `image`.
8+
9+
You need to provide an Ethereum wallet private key in `DKN_WALLET_SECRET_KEY` environment variable. The walelt address derived from this key represents your identity in Dria network.
10+
11+
You can also modify models supported by your node in `DKN_MODELS` environment variable. Consult the [guide](https://firstbatch.notion.site/How-to-Run-a-Node-ed2bef2c8eec4dd280286f2e081e51d2) to see the options.

dria/deploy.yaml

Lines changed: 95 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,95 @@
1+
---
2+
version: "2.0"
3+
4+
services:
5+
dkn:
6+
image: firstbatch/dkn-compute-node:master-3fe904d-1724095365
7+
expose:
8+
- port: 80
9+
as: 80
10+
to:
11+
- global: true
12+
env:
13+
# You need to provide a wallet private key here
14+
- DKN_WALLET_SECRET_KEY=
15+
# You can modify list of supported/enabled models - check the resources, might need a bump with more/different models
16+
- DKN_MODELS=,llama3.1:latest
17+
18+
# You can provide API keys to integrate with other services
19+
- OPENAI_API_KEY=
20+
- SERPER_API_KEY=
21+
- BROWSERLESS_TOKEN=
22+
- ANTHROPIC_API_KEY=
23+
24+
25+
# You probably don't want to touch these
26+
- OLLAMA_AUTO_PULL=true
27+
- OLLAMA_HOST=http://ollama
28+
- OLLAMA_PORT=11434
29+
- RUST_LOG=none,dkn_compute=info
30+
31+
- DKN_ADMIN_PUBLIC_KEY=0208ef5e65a9c656a6f92fb2c770d5d5e2ecffe02a6aade19207f75110be6ae658
32+
33+
ollama:
34+
image: ollama/ollama@sha256:532b563c005c1d60cf959d498418e0d3b5cc1f02f6a1608ae7d0c0a8a3c8a2f5 #this is the image `latest` pointed to on 2024-08-19, update if needed
35+
expose:
36+
- port: 11434
37+
as: 11434
38+
to:
39+
- global: false
40+
- service: dkn
41+
params:
42+
storage:
43+
ollamadata:
44+
mount: /root/.ollama
45+
readOnly: false
46+
profiles:
47+
compute:
48+
dkn:
49+
resources:
50+
cpu:
51+
units: 2
52+
memory:
53+
size: 2Gi
54+
storage:
55+
- size: 1Gi
56+
ollama:
57+
resources:
58+
cpu:
59+
units: 4
60+
memory:
61+
size: 10Gi
62+
storage:
63+
- size: 10Gi
64+
- name: ollamadata
65+
size: 24Gi
66+
attributes:
67+
persistent: true
68+
class: beta3
69+
gpu:
70+
units: 1
71+
attributes:
72+
vendor:
73+
nvidia:
74+
- model: rtx4090
75+
ram: 24Gi
76+
interface: pcie
77+
placement:
78+
dcloud:
79+
pricing:
80+
dkn:
81+
denom: uakt
82+
amount: 100000
83+
ollama:
84+
denom: uakt
85+
amount: 100000
86+
deployment:
87+
dkn:
88+
dcloud:
89+
profile: dkn
90+
count: 1
91+
ollama:
92+
dcloud:
93+
profile: ollama
94+
count: 1
95+

0 commit comments

Comments
 (0)