Ettore Di Giacinto 7809c5f5d0 fix(vision): propagate mtmd media marker from backend via ModelMetadata (#9412) ... Upstream llama.cpp (PR #21962) switched the server-side mtmd media marker to a random per-server string and removed the legacy "<__media__>" backward-compat replacement in mtmd_tokenizer. The Go layer still emitted the hardcoded "<__media__>", so on the non-tokenizer-template path the prompt arrived with a marker mtmd did not recognize and tokenization failed with "number of bitmaps (1) does not match number of markers (0)". Report the active media marker via ModelMetadataResponse.media_marker and substitute the sentinel "<__media__>" with it right before the gRPC call, after the backend has been loaded and probed. Also skip the Go-side multimodal templating entirely when UseTokenizerTemplate is true — llama.cpp's oaicompat_chat_params_parse already injects its own marker and StringContent is unused in that path. Backends that do not expose the field keep the legacy "<__media__>" behavior.		2026-04-18 20:30:13 +02:00
..
grpc	fix: speedup git submodule update with --single-branch (#2847)	2024-07-13 22:32:25 +02:00
ik-llama-cpp	chore: ⬆️ Update ikawrakow/ik_llama.cpp to 52efa12fdae390d1dca6ecd7ca00010fe51f651e (#9404)	2026-04-18 09:21:32 +02:00
llama-cpp	fix(vision): propagate mtmd media marker from backend via ModelMetadata (#9412)	2026-04-18 20:30:13 +02:00
turboquant	chore: ⬆️ Update TheTom/llama-cpp-turboquant to `45f8a066ed5f5bb38c695cec532f6cef9f4efa9d' (#9385)	2026-04-17 08:12:21 +02:00