Inference Node — runs on each GPU machine, cuBLAS SGEMM via Panama FFI, pipeline forward pass, gRPC activation transfer

Compile Dependencies (11)

Category/LicenseGroup / ArtifactVersionUpdates
Apache 2.0
cab.ml » api
Client-facing OpenAPI 3.0 REST spec + generated Java code.
0.1.1
Apache 2.0
cab.ml » lora
LoRA adapter tensors, checkpoint I/O, and Adam — no CUDA or
0.1.1
Apache 2.0
cab.ml » registry
Model Registry and Shard Loader — GGUF weight loading,
0.1.1
Apache 2.0
cab.ml » kvcache
KV Cache Manager — three-tier GPU/CPU/Disk storage, prefix Trie
0.1.1
Apache 2.0
cab.ml » health
Health Monitor — Hazelcast membership events, GPU probes (via
0.1.1
Cache Impl Apache 2.0
com.hazelcast » hazelcast
Core Hazelcast Module
5.4.05.7.0
Dist Communication Apache 2.0
io.grpc » grpc-netty-shaded1 vulnerability
gRPC: Netty Shaded
1.68.01.84.0
Serializer Apache 2.0
io.grpc » grpc-protobuf
gRPC: Protobuf
1.68.01.84.0
Dist Communication Apache 2.0
io.grpc » grpc-stub
Stub classes for GRPC which are exposed to developers and provides type-safe bindings.
1.68.01.84.0
io.micrometer » micrometer-core1 vulnerability
Core module of Micrometer containing instrumentation API and implementation
1.13.01.17.1
io.micrometer » micrometer-registry-prometheus
MeterRegistry implementation for Prometheus using io.prometheus:prometheus-metrics-core. If you have compatibility ...
1.13.01.17.1