HIP: build eatch ci build test for a different architecture (#21337)

fix: add openssl to nix dependencies (#21353) (#21355)

ggml-zendnn : add MUL_MAT_ID op support for MoE models (#21315)

vocab: fix Gemma4 tokenizer (#21343)

rpc : reuse compute graph buffers (#21299)

chat : avoid including json in chat.h (#21306)

(revert) kv-cache : do not quantize SWA KV cache (#21332)

ci : add AMD ZenDNN label to PR labeler (#21345)

[HIP] Bump ROCm version to 7.2.1 (#21066)

fix: gemma 4 template (#21326)

tests : add unit test coverage for llama_tensor_get_type (#20112)

ggml-webgpu: add vectorized flash attention (#20709)

tests: allow exporting graph ops from HF file without downloading weights (#21182)

model, mtmd: fix gguf conversion for audio/vision mmproj (#21309)

common : add commentary rules for gpt-oss-20b (#21286)