LvLLM is a special NUMA extension of vllm that makes full use of CPU and memory resources, reduces GPU memory requirements, and features an efficient GPU parallel and NUMA parallel architecture, supporting hybrid inference for MOE large models.
Adaptive disaggregated inference on a role-free fleet.
a drop-in replacement fork of the original bincode crate: "A binary serialization / deserialization strategy for transforming structs into bytes and vice versa!"
Yet another codec library for Go named "godec"
🛠️ Implement TOON in PHP for efficient serialization of JSON-like data, optimizing parsing for Large Language Models while maintaining clarity and structure.
Schemaless Protobuf inspection toolkit for .NET. Decodes raw wire-format payloads with no .proto files, generated classes, or schemas - auto-intercepting HTTP and gRPC (unary + streaming) traffic into human-readable trees or JSON for diagnostics, auditing, and reverse engineering.
🔍 PHP port of 💡 sequin — ANSI escape-sequence inspector: decodes SGR, CSI, OSC, DCS, APC, SS3 into human-readable labels.
One-command honest local-inference battery: /tokenize-true prompt sizing, ambient A/B noise control, spec-dec disclosure, machine-readable JSON
Ballerina encoding module.
A Deobfuscator for pyobfuscate.com
To associate your repository with the decode topic, visit your repo's landing page and select "manage topics."