build : remove GGML_METAL_USE_BF16 from all build scripts (ggml-org#2…

…6604)

sycl : support dev2dev memcpy by DEV2DEV_MEMCPY_FORWARD (ggml-org#26234)

Co-authored-by: Neo Zhang Jianyu <jianyu.zhang@intel.com>

ui: Add MCP Servers Opt-In for first time visitors (ggml-org#25239)

* feat: ui: Add predefined recommended MCP servers to settings

* feat: ui: Add MCP server recommendation dialog with custom server support

* feat: Auto-focus input fields on mount and dynamic addition

* feat: Add header validation to MCP server add and edit forms

* feat: Persist recommended MCP server opt-in selections

* test: Cover MCP configuration with tests

* chore: Format & cleanup

* feat: Centralize MCP server overrides to settings config and improve recommendation UI

* fix: Capture index before mutation to prevent focus drift

* refactor: Extract MCP_CARD_VISIBLE_TOOL_LIMIT to shared constants

* refactor: Support arbitrary authorization header schemes

* refactor: Consolidate MCP recommendations dismissal into existing storage key

* fix: Use case-insensitive comparison for MCP server ID prefix check

* refactor: Centralize MCP server visibility logic and extract recommendations hook

* refactor: Cleanup

mtmd: build_vit batching (ggml-org#24352)

vocab : support tokenizer for LFM2.5-8B-A1B (ggml-org#23826)

* vocab: Support tokenizer for LFM2.5-8B-A1B

* Keep liquid6 tokenizer in models

ci : releases use Github-hosted builds for the UI (ggml-org#23823)

* ci : releases use Github-hosted builds for the UI

* cont : fix name

vulkan: Switch MUL_MAT_VEC to 4 K per iteration for F16/32 (ggml-org#…

…22887)

* vulkan: Switch MUL_MAT_VEC to 4 K per iteration for F16/32

Against mesa git, this shows a 4.8% performance improvement for
tg128 on Qwen3.5-9B:BF16 on Intel BMG.

Note that this breaks some tests until the last commit which fixes
OOB A reads.

* vulkan: Use aligned loads in mul_mat_vec when available

Against mesa git, this shows a 3.3% performance improvement for
tg128 on Qwen3.5-9B:BF16 on Intel BMG.

* Make explicit that `num_rows` is <= `NUM_ROWS` in mul_mat_vec

Mesa's UUB logic can't see through conditionals, limiting its
ability to understand the bounds on the `num_rows` field in the
cleanup run. Making it explicit that `num_rows` is, indeed, always
<= `NUM_ROWS` helps mesa make slightly better codegen.

Against mesa git, this currently shows a 1% performance improvement
in tg128 on Qwen3.5-9B:BF16 on Intel BMG.

* vulkan: Fix OOB A reads in MUL_MAT_VEC for odd sizes

There was a TODO to fix the OOB reads from the A matrix which we do
here.

It is within performance noise (+<0.1%) in tg128 for
Qwen3.5-9B:BF16 on Intel BMG.

convert : support Gemma4ForCausalLM architecture (ggml-org#23682)

* convert : support Gemma4ForCausalLM architecture (ggml-org#23674)

* fix indent

---------

Co-authored-by: Oleg Afonin <your.email@example.com>
Co-authored-by: Sigbjørn Skjæret <sigbjorn.skjaeret@scala.com>

model: tag ffn_latent as MUL_MAT to fix buft probe (ggml-org#23664)

ffn_latent_down/up are declared GGML_OP_MUL in LLM_TENSOR_INFOS but
nemotron-h feeds them through ggml_mul_mat. The loader buft probe asks
the backend about the declared op, so it tested an elementwise MUL on a
q8_0 weight. That used to return true unconditionally and the weight
stayed on GPU by luck. Once supports_op told the truth, the probe got a
no and the loader pushed the weight and its matmul to CPU, splitting the
graph. Tagging it MUL_MAT asks the real question, the math is unchanged.

Verified on Nemotron 3 Super 120B Q5_K_M: from 64.9 back to 103.22 t/s.

requirements : bump torch to 2.11.0 (ggml-org#23503)

* requirements: relax torch~=2.6.0 to torch>=2.6.0 for convert_hf_to_gguf

The ~=2.6.0 operator resolves to >=2.6.0, <2.7.0, which fails on
PyPI for platform/CPython combinations where 2.6.x is not present.
The accompanying comment already says 'PyTorch 2.6.0 or later', so
the looser >=2.6.0 matches the documented intent and unblocks
pip install -r requirements/requirements-convert_hf_to_gguf.txt.

Fixes ggml-org#23408

* requirements: bump torch floor to 2.11.0 per maintainer

* requirements: pin torch to ==2.11.0 per project policy

* requirements: pin mtmd torch and torchvision to 2.11.0/0.26.0 per project policy

* requirements: suppress check_requirements pin warning on mtmd

The check_requirements script flags '==' on lines in files matched by
*/**/requirements*.txt. Append the documented suppression comment to the
pinned torch and torchvision lines (and to the s390x platform marker lines)
so the check passes while keeping the pins required by project policy.

* ty: silence Tensor/Module union check on model[0].auto_model

With torch 2.11.0 stubs, nn.Sequential.__getitem__ now returns
Tensor | Module rather than Module, so model[0].auto_model fails ty
on the SentenceTransformer code path. The runtime behavior is
unchanged because SentenceTransformer always wraps a Module at
index 0. Adding a targeted unresolved-attribute ignore keeps the
type-check green without altering behavior. A follow-up issue
tracks typing the variable explicitly.