* CI: fix ARM64 image build error & enable compilation

* Update .github/workflows/docker.yml

Co-authored-by: Aaron Teo <taronaeo@gmail.com>

* CI: revert ggml/src/ggml-cpu/CMakeLists.txt

* Update .github/workflows/docker.yml

Co-authored-by: Aaron Teo <taronaeo@gmail.com>

* CI: update runs-on to ubuntu24.04, and update ARM64 build image ( ubuntu_version: "24.04")

* CI: change cpu.Dockerfile gcc to 14;

* CI : cpu.Dockerfile , update pip install .

* Update .github/workflows/docker.yml

Co-authored-by: Aaron Teo <taronaeo@gmail.com>

---------

Co-authored-by: Aaron Teo <taronaeo@gmail.com>

* add /glob command

* output error when max files reached

* support globbing outside curdir

…ml-org#21085)

* fix whitespace reasoning issues + add reconstruction tests

* Proper fix

* fix Nemotron autoparser test expectations to include newline in marker

* vulkan: add noncontiguous GLU support

* fix compile issue

Signed-off-by: Adrien Gallouët <angt@huggingface.co>

* refactor: Make `DialogConfirmation` extensible with children slot

* feat: Add conversation forking logic

* feat: Conversation forking UI

* feat: Update delete/edit dialogs and logic for forks

* refactor: Improve Chat Sidebar UX and add MCP Servers entry

* refactor: Cleanup

* feat: Update message in place when editing leaf nodes

* chore: Cleanup

* chore: Cleanup

* chore: Cleanup

* chore: Cleanup

* chore: Cleanup

* chore: Cleanup

* refactor: Post-review improvements

* chore: update webui build output

* test: Update Storybook test

* chore: update webui build output

* chore: update webui build output

…19771)

Signed-off-by: Adrien Gallouët <angt@huggingface.co>

…g#21107)

…schema pattern converter (ggml-org#21124)

The regex-to-grammar converter in _visit_pattern() crashes with SIGSEGV
when a JSON schema "pattern" field contains a non-capturing group (?:...).

Root cause: when the parser sees '(' followed by '?', it pushes a warning
but does not advance past '?:'. The recursive transform() call then
interprets '?' as a quantifier and calls seq.back() on an empty vector,
causing undefined behavior.

This commonly occurs when serving OpenAI-compatible tool calls from
clients that include complex regex patterns in their JSON schemas (e.g.,
date validation patterns like ^(?:(?:\d\d[2468][048]|...)-02-29|...)$).

The fix:
- Skip '?:' after '(' to treat non-capturing groups as regular groups
- For unsupported syntax (?=, ?!, etc.), skip to matching ')' safely,
  handling escaped characters to avoid miscounting parenthesis depth
- Adjust the ')' unbalanced-parentheses check using direct char
  comparisons instead of substr
- Add test cases for non-capturing groups (C++ only, as the JS/Python
  implementations do not yet support this syntax)

* remove/replace nested button elements

* map rest props to outer element

* solve TODO

* chore: update webui build output

* add character class support to glob_match

* remove pointless reference

…s key (ggml-org#21128)

…ggml-org#21093)

* use half cores to build, avoid OS hang

* reduce the output text num to short test time

* avoid to return 0

* hex-fa: add simple dma cache for Mask

I noticed that we were refetch the mask rows over and over.
This simple cache avoids that.

* hex-dma: unset in-order desc bit which caused signficant perf regression

We don't rely on true in order processing of the DMA descriptors anywhere.
Turns out this mode caused significant regression of around 3-4 TPS during token gen.

* hex-rope: update comment to clarify that we don't need in-order DMA completions

* Optimize MOE GEMV kernel for BS > 1.

The previous MOE kernel for BS > 1 had too many thread blocks (nrows_x, nchannels_dst, ncols_dst), with very little work per block. block of (32, 4) was doing inner dot product for a single row.

New mul_mat_vec_q_moe kernel is dedicated for MoE multi-token kernel with grid (ceil(nrows_x/rpb), nchannels_dst), block (warp_size, ncols_dst). Each warp handles two rows independently with warp-level reduction only (no shared memory sync).

This change doesn't increase any compilation time as a single template instance is needed per type. This also simplifies the original GEMV kernel and gets rid of `is_multi_token_id` specialization.

* Remove em-dashes

* Cherry-pick changes from @am17an PR ggml-org#20885 to enable small_k optimization only for cases where it benefits

Increase max batch size for MMVQ kernels for MUL_MAT_ID to 8

* Make the max batch size for MOE GEMV kernel configurable based on GPU arch and datatype

---------

Co-authored-by: Aman Gupta <amangupta052@gmail.com>

* server: wrap headers for mcp proxy

* Update tools/server/server-cors-proxy.h

Co-authored-by: Georgi Gerganov <ggerganov@gmail.com>

* fix build

* chore: update webui build output

* chore: update webui build output

---------

Co-authored-by: Georgi Gerganov <ggerganov@gmail.com>
Co-authored-by: Aleksander Grygier <aleksander.grygier@gmail.com>

* fix incorrect type ignore comments

* bump ty to 0.0.26

…l-org#20978)

* llama-model-loader: use pinned memory for tensor overrides

* change to warning

* fix: Branching logic + small refactor

* chore: update webui build output

When RPC is running with a remote backend which doesn't have init_tensor
function (like CPU and Metal), the server log gets full with error
messages saying that init_tensor is being called with null buffer which
is incorrect. This patch fixes this.

…l-org#21181)

* CUDA: Fix CUB's argsort when nrows % block_size == 0 CCCL < 3.1

We wrongly calculated offset_grid as `ceildiv(nrows, block_size)`,
while it must be `ceildiv(nrows + 1, block_size)`. As a consequence, we
had uninitialized values in `offset_iterator[nrows]` for the case when
`nrows % block_size == 0`.

Fixes ggml-org#21162

* Reduce nrows in test case to 256, don't need 768

* Reject empty computed member expressions before returning slices[0] from parse_member_expression_arguments().

* Treat empty computed member expressions with Jinja2 undefined semantics

Treat empty computed member expressions like `a[]` as undefined instead of
raising a parser error, to match Jinja2 behavior.

- return a noop expression for empty computed member arguments
- return undefined when a computed member key evaluates to undefined
- add Jinja tests covering `a[]|default('fallback')` and `a[] is undefined`

* Handle undefined computed member properties

Move undefined-property handling to the common member access path, and add a test covering `a[undefined] is undefined`.

* Use default undefined value in member access

Initialize val and then return it when property is undefined.

Co-authored-by: Sigbjørn Skjæret <sigbjorn.skjaeret@scala.com>

* empty statement parses to blank_expression instead of noop_statement

---------

Co-authored-by: Sigbjørn Skjæret <sigbjorn.skjaeret@scala.com>

* CI: Enable CUDA and Vulkan ARM64 runners and fix CI/CD

Co-authored-by: Ts-sound <44093942+Ts-sound@users.noreply.github.com>

* Obtain source tag name from git tag

Co-authored-by: Sigbjørn Skjæret <sigbjorn.skjaeret@scala.com>

---------

Co-authored-by: Ts-sound <44093942+Ts-sound@users.noreply.github.com>
Co-authored-by: Sigbjørn Skjæret <sigbjorn.skjaeret@scala.com>

* opencl: add q4_K gemm and gemv kernels for Adreno

* opencl: fix whitespace

* opencl: add workarounds for compiler bugs on older devices

* opencl: handle fp16 denorm on X Elite

* opencl: fix kernel build error

* opencl: fix whitespace

* opencl: make q4_K cvt kernels signature consistent

---------

Co-authored-by: Li He <lih@qti.qualcomm.com>

…#21337)

This helps improve our chances of finding build failures before the release workflow
builds for all architectures.

…rg#21331)

The `HSA_OVERRIDE_GFX_VERSION` variable can be used in ROCm to override an unsupported target architecture with a similar but supported target architecture.

This does not and has never worked on Windows. I think the clarification could avoid driving Windows people towards this solution that does not work.

Co-authored-by: M1DNYT3 <m1dnyt3@MacBookPro.lan>
Co-authored-by: CISC <CISC@users.noreply.github.com>

…gml-org#21327)

* common : fix tool call type detection for nullable and enum schemas

* common, tests : fix grammar delegation for nullable/enum schemas and add tests

Fix enum type inference to scan all enum values (not just index 0) so
schemas like {"enum": [0, "celsius"]} correctly detect string type.

Fix schema_delegates in peg-parser to handle nullable type arrays
(["string", "null"]) and typeless enum schemas in raw mode, allowing
the tagged parser to use raw text instead of JSON-formatted strings.

Add test cases for Qwen3-Coder (TAG_WITH_TAGGED format):
- nullable string ["string", "null"]
- nullable string with null first ["null", "string"]
- nullable integer ["integer", "null"]
- enum without explicit type key

…ity for tag-json parsers (ggml-org#21230)

* Fix call ID detection (Mistral parser mostly) + atomicity for tag-json parsers

* Rename

* Update common/chat-auto-parser-generator.cpp

Co-authored-by: Sigbjørn Skjæret <sigbjorn.skjaeret@scala.com>

---------

Co-authored-by: Sigbjørn Skjæret <sigbjorn.skjaeret@scala.com>

…org#20993)

* server: clear idle slots KV from VRAM (LLAMA_KV_KEEP_ONLY_ACTIVE)

* server: move idle slot KV clearing to slot release

The save "cost" is now paid by the finishing request.

* server: add --kv-clear-idle flag, enable by default

* server: skip clearing last idle slot, clear on launch

* server: test --no-kv-clear-idle flag

* server: simplify on-release clearing loop

* server: remove on-release KV clearing, keep launch-only

* cont : clean-up

* tests: update log strings after --clear-idle rename

* tests: use debug tags instead of log message matching

* test: fix Windows CI by dropping temp log file unlink

---------

Co-authored-by: Georgi Gerganov <ggerganov@gmail.com>

* experimenting CI

* Experimenting CI fix for MinGW

* experimenting CI on Windows

* modified script for integration with VisualStudio

* added proxy handling

* adding python version for Windows execution

* fix iterator::end() dereference

* fixed proxy handling

* Fix errors occurring on Windows

* fixed ci script

* Reverted to master

* Stripping test items to simplify Windows test

* adjusting script for windows testing

* Changed shell

* Fixed shell

* Fixed shell

* Fix CI setting

* Fix CI setting

* Fix CI setting

* Experimenting ci fix

* Experimenting ci fix

* Experimenting ci fix

* Experimenting ci fix

* experimenting fix for unit test error

* Changed to use BUILD_LOW_PERF to skip python tests

* Fix CI

* Added option to specify Ninja generator

* Reverted proxy related changes

…fsets (ggml-org#21278)

* Work towards removing bitcast

* Move rest of existing types over

* Add timeout back to wait and remove synchronous set_tensor/memset_tensor

* move to unpackf16 for wider compatibility

* cleanup

* Remove deadlock condition in free_bufs

* Start work on removing parameter buffer pools

* Simplify and optimize further

* simplify profile futures

* Fix stride

* Try using a single command buffer per batch

* formatting

…-sm121

# Conflicts:
#	ggml/src/ggml-cuda/fattn.cu
#	src/llama-graph.cpp

…rg#21038)

Upstream master now applies Walsh-Hadamard rotation to K/V/Q before KV
cache storage (commit 744c0c7). This is the same rotation TBQ was
doing independently, causing double rotation after the merge.

TBQ types are now pure codebook quantizers:
- SET_ROWS: normalize + codebook quantize + pack (no FWHT)
- FA dequant: codebook lookup + scale (no Q pre-rotation, no V inverse rotation)
- Standalone dequant: codebook lookup + scale (no inverse FWHT)

Removes ~200 lines of rotation code from CUDA and CPU paths.
Fixes garbage output caused by double WHT rotation after upstream merge.