66 lines
1.4 KiB
TOML
66 lines
1.4 KiB
TOML
|
[taskserv]
|
||
|
|
name = "vapora-llm-router"
|
||
|
|
type = "service"
|
||
|
|
version = "0.2.0"
|
||
|
|
description = "Multi-IA Router - Route tasks to optimal LLM provider"
|
||
|
|
|
||
|
|
[source]
|
||
|
|
repository = "ssh://git@repo.jesusperez.pro:32225/jesus/Vapora.git"
|
||
|
|
branch = "main"
|
||
|
|
path = "vapora-llm-router/"
|
||
|
|
|
||
|
|
[build]
|
||
|
|
runtime = "rust"
|
||
|
|
build_command = "cargo build --release -p vapora-llm-router"
|
||
|
|
|
||
|
|
[deployment]
|
||
|
|
namespace = "vapora-system"
|
||
|
|
replicas = 2
|
||
|
|
image = "vapora/llm-router"
|
||
|
|
image_tag = "${version}"
|
||
|
|
|
||
|
|
[ports]
|
||
|
|
http = 8899
|
||
|
|
metrics = 9090
|
||
|
|
|
||
|
|
[resources]
|
||
|
|
requests = { cpu = "1000m", memory = "2Gi" }
|
||
|
|
limits = { cpu = "2000m", memory = "4Gi" }
|
||
|
|
|
||
|
|
[health_check]
|
||
|
|
type = "http"
|
||
|
|
path = "/health"
|
||
|
|
interval_secs = 10
|
||
|
|
|
||
|
|
[environment]
|
||
|
|
DATABASE_URL = "surrealdb://surrealdb-0.vapora-system:8000"
|
||
|
|
REDIS_URL = "redis://redis-0.vapora-system:6379"
|
||
|
|
ROUTING_MODE = "hybrid" # hybrid | static | dynamic
|
||
|
|
COST_TRACKING_ENABLED = "true"
|
||
|
|
RUST_LOG = "debug,vapora_llm_router=trace"
|
||
|
|
|
||
|
|
[secrets]
|
||
|
|
CLAUDE_API_KEY = "secret:vapora-secrets:claude-api-key"
|
||
|
|
OPENAI_API_KEY = "secret:vapora-secrets:openai-api-key"
|
||
|
|
GEMINI_API_KEY = "secret:vapora-secrets:gemini-api-key"
|
||
|
|
|
||
|
|
[config_maps]
|
||
|
|
ROUTING_RULES = "configmap:vapora-routing-rules"
|
||
|
|
LLM_MODELS = "configmap:vapora-llm-models"
|
||
|
|
|
||
|
|
[scaling]
|
||
|
|
min_replicas = 2
|
||
|
|
max_replicas = 5
|
||
|
|
target_cpu = 60
|
||
|
|
|
||
|
|
[persistence]
|
||
|
|
enabled = true
|
||
|
|
size = "5Gi"
|
||
|
|
mount_path = "/routing-cache"
|
||
|
|
storage_class = "ssd"
|
||
|
|
|
||
|
|
[monitoring]
|
||
|
|
prometheus_metrics = true
|
||
|
|
track_costs = true
|
||
|
|
track_latencies = true
|