bots-garden/houlapublic Fork
⑂ v0.0.1
Commits
⬇ Clone ▾
git clone https://git.rickub.com/bots-garden/houla.git
git clone ssh://git@rickub.com/bots-garden/houla.git

Host key fingerprint (ed25519): SHA256:iycHnxEyq0Q7uyVpB7JlznP0G7JrTPXLYRcAU5CSLhc — verify it before your first connect.

Download ZIP
🛟 Updated. 1dfc968Unverified · on v0.0.1 · k33g · 4h ago
llama-cpp-olala.yaml · 29 lines · 1.0 KBYAML Blame HistoryRaw
 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
# syntax=docker/sandbox-kit:3
schemaVersion: "3"
kind: mixin
displayName: llama.cpp (OVHai Olala fork, CPU)
description: llama-server and llama-cli from the OVHai llama.cpp fork, which supports the `olala` architecture (CPU only)

args:
  ref:
    # Commit of https://github.com/OVHai-LLM/llama.cpp to build. Pin a full SHA for reproducible builds.
    default: "55c22dca0c987f247c21b427c3b3a09e1a28187f"
    pattern: '^[0-9a-f]{40}$'
    buildArg: LLAMA_REF

# provides needs an X.Y.Z version (a commit SHA is rejected by the kit frontend):
# 0.5.0 = what the fork reports (`llama-server --version`: 0.5.0-dev). The commit is pinned by `ref`.
provides: ["llama-cpp-olala@0.5.0"]

capabilities:
  # Runtime access to download GGUF models from Hugging Face (hf download, llama-server -hf).
  - type: com.docker.sandbox/network-policy@1
    config:
      runtime:
        allow:
          - huggingface.co
          - "*.huggingface.co"
          - "*.hf.co"
  - type: com.docker.sandbox/agent-context@1
    config:
      contentFile: ./llama-cpp-olala-context.md