parallel : add option for different RNG seeds (#14757)
ggml-ci

cuda : Fix Gemma3n not executed as CUDA_GRAPH on NVGPUs (#14741)
* Fix Gemma3n not executed as CUDA_GRAPH on NVGPUs
Gemma3n uses Matrix-Matrix addition as part of their input processing,
wrongly triggering CUDA_GRAPH disablement on NVGPUs even when batch-size
of 1 is used.
* Exclude `project_per_layer_input` by matching node names
This ensures that all other graphs which don't exhibit this pattern do
not have their behavior changed.
* Revert unnecessary formatting changes

graph : avoid huge warm-up graphs for MoE models (#14753)
* graph : avoid huge warm-up graphs for MoE models
ggml-ci
* cont : bump max nodes to 8x model tensors

model : fix build after merge conflict (#14754)

CUDA: set_rows + cpy.cu refactor (#14712)

graph : refactor context to not pass gf explicitly (#14629)
ggml-ci

graph : Pass the graph placeholder message in debug mode (#14748)
Without that condition, this debug log clutters the screen every batch treated in the prompt processing, or every token generated in Kobold.cpp.

use max work group size for device to replace the magic number (#14732)

model: add Ernie 4.5 MoE support (#14658)
* Add Ernie4.5 MoE
* Fix Flake errors.
* Properly encode/decode MoE layer step
* Correct tensor mappings (.weight)
* Pass and read n_ff_exp
* n_ff_shexp calculation and further minor changes
* Rope fixes.
* .gitignore fix
* Add unit32 cast for Linux builds
* Apply suggestions from code review
Co-authored-by: Sigbjørn Skjæret <sigbjorn.skjaeret@scala.com>
* Further fixes from code review
* Fix trailing whitespace
* Reenable missing experts error
* Code style from code review
Co-authored-by: Sigbjørn Skjæret <sigbjorn.skjaeret@scala.com>
* Fix non-MoE regression
Co-authored-by: Sigbjørn Skjæret <sigbjorn.skjaeret@scala.com>
---------
Co-authored-by: Sigbjørn Skjæret <sigbjorn.skjaeret@scala.com>

kv-cache : fix k-shift for multiple streams (#14742)
ggml-ci