UPSTREAM PR #21835: Expose build_info in router mode

[loci-dev]

Note

Source pull request: ggml-org/llama.cpp#21835

Overview

• Include the build_info in llama-server 's /prop endpoint when running in router mode;
• also expose params.models_max instead of dummy value and params.models_autoload

Requirements

• I have read and agree with the contributing guidelines
• AI usage disclosure: YES (code review)

---

For context, with this change, in router mode, for GET /props

BEFORE

{
  "role":"router",
  "max_instances":4,
  "model_alias":"llama-server",
  "model_path":"none",
  "default_generation_settings":{"params":null,"n_ctx":0},
  "webui_settings":{}
}

AFTER

{
  "role":"router",
  "max_instances":4,
  "models_autoload":true,
  "model_alias":"llama-server",
  "model_path":"none",
  "default_generation_settings":{"params":null,"n_ctx":0},
  "webui_settings":{},
  "build_info":"b8771-9ef1fab9c"
}

[loci-review]

No meaningful performance changes were detected across 127140 analyzed functions in the following binaries: build.bin.llama-cvector-generator, build.bin.llama-bench, build.bin.libllama.so, build.bin.libmtmd.so, build.bin.llama-tts, build.bin.llama-quantize, build.bin.llama-qwen2vl-cli, build.bin.llama-tokenize, build.bin.llama-gemma3-cli, build.bin.llama-gguf-split, build.bin.llama-llava-cli, build.bin.llama-minicpmv-cli, build.bin.libggml.so, build.bin.libggml-cpu.so, build.bin.libggml-base.so.

💬 Questions? Tag @loci-dev