ai-workspace-infra/playbooks
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

feat(litellm): update model registrations and gateway configurations with API key gating

Browse Source
This commit is contained in:
Haitao Pan 2026-06-23 11:03:25 +08:00
parent 01f1499a60
commit e7c96675ff
3 changed files with 20 additions and 83 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

62
roles/vhosts/gateway_openclaw/defaults/main.yml
View File

@ -62,15 +62,9 @@ gateway_openclaw_default_model:
primary: "{{ gateway_openclaw_default_model_primary }}"
fallbacks:
- "{{ gateway_openclaw_default_model_fallback }}"
- nvidia/nemotron-3-super-120b-a12b
- nvidia/minimaxai/minimax-m2.5
- nvidia/z-ai/glm5
gateway_openclaw_default_models:
"{{ gateway_openclaw_default_model_primary }}": {}
"{{ gateway_openclaw_default_model_fallback }}": {}
nvidia/nemotron-3-super-120b-a12b: {}
nvidia/minimaxai/minimax-m2.5: {}
nvidia/z-ai/glm5: {}
openai/gpt-5.5:
agentRuntime:
id: codex
@ -139,59 +133,3 @@ gateway_openclaw_model_providers:
contextWindow: 128000
maxTokens: 8192
reasoning: true
nvidia:
api: openai-completions
baseUrl: https://integrate.api.nvidia.com/v1
models:
- id: nvidia/nemotron-3-super-120b-a12b
name: NVIDIA Nemotron 3 Super 120B
input: [text]
contextWindow: 262144
maxTokens: 8192
reasoning: false
compat:
requiresStringContent: true
cost:
input: 0
output: 0
cacheRead: 0
cacheWrite: 0
- id: moonshotai/kimi-k2.5
name: Kimi K2.5
input: [text]
contextWindow: 262144
maxTokens: 8192
reasoning: false
compat:
requiresStringContent: true
cost:
input: 0
output: 0
cacheRead: 0
cacheWrite: 0
- id: minimaxai/minimax-m2.5
name: MiniMax M2.5
input: [text]
contextWindow: 196608
maxTokens: 8192
reasoning: false
compat:
requiresStringContent: true
cost:
input: 0
output: 0
cacheRead: 0
cacheWrite: 0
- id: z-ai/glm5
name: GLM-5
input: [text]
contextWindow: 202752
maxTokens: 8192
reasoning: false
compat:
requiresStringContent: true
cost:
input: 0
output: 0
cacheRead: 0
cacheWrite: 0

35
roles/vhosts/litellm/files/register_mainstream_models.sh
View File

@ -27,6 +27,12 @@ add_model() {
local api_key_env_var="$3"
local api_base="${4:-}"
# Skip registration when the backing API key was not provided (empty env var).
if [ -z "${!api_key_env_var:-}" ]; then
echo "[SKIP] $alias_name: $api_key_env_var is empty; not registering."
return 0
fi
echo "Adding model: $alias_name -> $litellm_provider_model"
local payload
@ -85,30 +91,15 @@ if [ -n "${DEEPSEEK_API_KEY:-}" ]; then
echo "========================================="
add_model "deepseek-v4-flash" "deepseek/deepseek-v4-flash" "DEEPSEEK_API_KEY"
add_model "deepseek-v4-pro" "deepseek/deepseek-v4-pro" "DEEPSEEK_API_KEY"
add_model "deepseek-chat" "deepseek/deepseek-chat" "DEEPSEEK_API_KEY"
add_model "deepseek-reasoner" "deepseek/deepseek-reasoner" "DEEPSEEK_API_KEY"
fi
if [ -n "${NVIDIA_API_KEY:-}" ]; then
echo "========================================="
echo "Registering NVIDIA Build Models..."
echo "========================================="
add_model "minimax-2.5" "openai/minimax/minimax-2.5" "NVIDIA_API_KEY" "https://integrate.api.nvidia.com/v1"
add_model "minimax-2.7" "openai/minimax/minimax-2.7" "NVIDIA_API_KEY" "https://integrate.api.nvidia.com/v1"
add_model "minimax-3.0" "openai/minimax/minimax-3.0" "NVIDIA_API_KEY" "https://integrate.api.nvidia.com/v1"
add_model "glm-4" "openai/thudm/glm-4-9b-chat" "NVIDIA_API_KEY" "https://integrate.api.nvidia.com/v1"
add_model "glm-5" "openai/thudm/glm-5" "NVIDIA_API_KEY" "https://integrate.api.nvidia.com/v1"
fi
if [ -n "${OLLAMA_API_KEY:-}" ]; then
echo "========================================="
echo "Registering OLLAMA/Proxy Models (Kimi, Qinwen, GLM)..."
echo "========================================="
OLLAMA_API_BASE="${OLLAMA_API_BASE:-https://api.ollama.cloud/v1}"
add_model "glm-5.x" "openai/glm-5" "OLLAMA_API_KEY" "$OLLAMA_API_BASE"
add_model "kimi" "openai/moonshot-v1-auto" "OLLAMA_API_KEY" "$OLLAMA_API_BASE"
add_model "qwen" "openai/qwen-max" "OLLAMA_API_KEY" "$OLLAMA_API_BASE"
add_model "ollama-llama3" "openai/llama3" "OLLAMA_API_KEY" "$OLLAMA_API_BASE"
add_model "nvidia/glm-5.2" "openai/thudm/glm-5.2-chat" "NVIDIA_API_KEY" "https://integrate.api.nvidia.com/v1"
add_model "nvidia/minimax-m3" "openai/minimax/minimax-m3" "NVIDIA_API_KEY" "https://integrate.api.nvidia.com/v1"
add_model "nvidia/qwen3.5" "openai/alibaba/qwen3.5-72b-instruct" "NVIDIA_API_KEY" "https://integrate.api.nvidia.com/v1"
fi
echo "========================================="
@ -138,5 +129,13 @@ if [ -n "${ANTHROPIC_API_KEY:-}" ]; then
add_model "claude-3-opus" "anthropic/claude-3-opus-20240229" "ANTHROPIC_API_KEY"
fi
if [ -n "${OLLAMA_API_KEY:-}" ]; then
echo "========================================="
echo "Registering OLLAMA Cloud Models..."
echo "========================================="
OLLAMA_API_BASE="${OLLAMA_API_BASE:-https://api.ollama.cloud/v1}"
add_model "ollama-cloud/kimi-k2.7-code" "openai/moonshot/kimi-k2.7-code" "OLLAMA_API_KEY" "$OLLAMA_API_BASE"
fi
echo "All models requested have been registered."
echo "You can check them at $LITELLM_URL/ui/?page=models"

2
setup-xworkspace-console.yaml
View File

@ -17,7 +17,7 @@
ai_workspace_prebuilt_components_required: "{{ lookup('ansible.builtin.env', 'AI_WORKSPACE_PREBUILT_COMPONENTS_REQUIRED') | default('false', true) | bool }}"
xworkspace_console_dashboard_dir: "{{ xworkspace_console_repo_dir }}/dashboard"
xworkspace_console_api_dir: "{{ xworkspace_console_repo_dir }}/api"
xworkspace_console_api_binary: "{{ xworkspace_console_repo_dir }}/bin/xworkspace-api"
xworkspace_console_api_binary: "{{ xworkspace_console_api_dir }}/xworkspace-api"
xworkspace_console_runtime_marker: "{{ xworkspace_console_repo_dir }}/.runtime-archive-sha256"
xworkspace_console_api_working_dir: "{{ xworkspace_console_repo_dir }}"
xworkspace_console_api_exec: "{{ xworkspace_console_api_binary }}"