UPSTREAM PR #19117: server: print actual model name in 'model not found" error

[loci-dev]

Note

Source pull request: ggml-org/llama.cpp#19117

Experimenting with AI, my environment gets messy fast and it's not always easy to know what model my software is trying to load. This helps with troubleshooting.

Before:

> Who are you ?


Error: {
  code = 400,
  message = "model not found",
  type = "invalid_request_error"
}

after

> Who are you ?


Error: {
  code = 400,
  message = "model 'toto' is not found",
  type = "invalid_request_error"
}

NB: I couldn't find a target to run linting (usually make lint / or make format) so tried to run the CI locally as explained in contributing.md with

GG_BUILD_CUDA=1 bash ./ci/run.sh ./tmp/results ./tmp/mnt
Warning: Using fallback CUDA architectures
./ci/run.sh: ligne 651: python3 : commande introuvable

I am developing from within the repo flake.nix so I expected all devtools present (I noticed the flake didn't provide ccache either, not sure if that's a conscious decision or not). I can submit a patch to add python3.

Make sure to read the contributing guidelines before submitting a PR

[loci-review]

No meaningful performance changes were detected across 115327 analyzed functions in the following binaries: build.bin.libllama.so, build.bin.llama-tts, build.bin.llama-cvector-generator, build.bin.libmtmd.so, build.bin.llama-tokenize, build.bin.llama-quantize, build.bin.llama-qwen2vl-cli, build.bin.libggml.so, build.bin.libggml-cpu.so, build.bin.libggml-base.so, build.bin.llama-bench, build.bin.llama-gguf-split, build.bin.llama-llava-cli, build.bin.llama-minicpmv-cli, build.bin.llama-gemma3-cli.

🔎 Full breakdown: Loci Inspector.
💬 Questions? Tag @loci-dev.