Commit Graph
11378 Commits
Author SHA1 Message Date
Adrien Gallouët edd6e2bbda common : add common_is_tty() helper and fix deprecated warnings on Windows (#29860)
Signed-off-by: Adrien Gallouët <angt@huggingface.co>
2026-10-03 15:48:24 +02:00
Yash Raj Pandey 9bf55f4a36 chat : honor json_schema in Ling 3.0 parser (#29813)
* chat : honor json_schema in Ling 3.0 parser

Ling 3.0 only built a grammar for tool calls and did not handle inputs.json_schema, so response_format requests were left unconstrained.

Add an eager response-format grammar path with precedence over tools, following the existing parser patterns. Require </think> before JSON when thinking is enabled and do not allow trailing prose after the JSON response.

Fixes #29652.

Assisted-by: Claude Opus 5.5

* chat : require Ling 3.0 think block for response formats
2026-10-03 15:40:50 +02:00
Pascal a55e952b85 ci: fix flaky ADD_ADD f16 by using the fused ADD tolerance (#29904) 2026-10-03 14:56:07 +02:00
Pascal 436f6f89e1 graph: gather the recurrent states once so the reserve covers every split (#29856)
build_rs gathered the extra states (n_rs - n_seqs rows) with their own
get_rows. The worst-case reserve has n_rs == n_seqs, so that node was
sized at zero rows, and any ubatch whose cells are not contiguous forced
a graph reallocation at an unchanged node count, which aborts under
GGML_SCHED_NO_REALLOC.

A single get_rows now gathers the n_rs states: the ubatch states and the
extra states are views of it, and its size only depends on n_rs, which
the reserve already sets to the maximum. A custom getter (mamba ssm_scan)
gathers from the second state, so a single sequence ubatch copies no
state. The views are built once per graph in the input to keep the host
overhead of the graph unchanged.
2026-10-03 14:02:05 +02:00
b92761a515 ggml-openvino: update to 2026.4.1, optimize performance, expand ops, improve device listing. (#29852)
* ggml-openvino : Qwen3.5 MoE perf (#312)

Squash of ravi9/llama.cpp#312:

- ggml-openvino: add detailed inference profiling (Yu, Zijun)
- ggml-openvino: use remote output tensors by default (Yu, Zijun)
- ggml-openvino: optimize single-sequence recurrent state (Yu, Zijun)
- opt1: remove recurrent reset for single sequence, opt2: direct gdn outputs (break parallel sequence) (Yu, Zijun)
- fix parallel sequences (Yu, Zijun)
- ggml-openvino: simplify graph cache key (ynimmaga)
- enable stateful for qwen35 single sequence (Yu, Zijun)
- Fix after rebasing (Yu, Zijun)
- Add k-requant option q4_asym64 (Yu, Zijun)
- Fix qwen35 llama-bench -p 0 (Yu, Zijun)
- Simplify RESHAPE translation (Yu, Zijun)
- openvino: fuse MoE routing (Yu, Zijun)
- openvino: fuse GDN qk normalization (Yu, Zijun)
- openvino: enable GPU MoE fusion by default (Yu, Zijun)
- ggml-openvino: add cache_only mode to import cached compiled model on disk directly (Yu, Zijun)
- openvino : report the device allocation limit to ggml (Łukasz Ślusarczyk)
- Fix windows build (Yu, Zijun)

Co-authored-by: ynimmaga <ynimmaga@users.noreply.github.com>
Co-authored-by: Łukasz Ślusarczyk <lukasz.slusarczyk@intel.com>

* ggml-openvino: Update doc of compiled model cache

* openvino: implement PRD-compliant device enumeration and memory reporting

* openvino: fix multi-device listing issues from review

- Only the device selected by GGML_OPENVINO_DEVICE reports as GPU; the
  other OpenVINO devices report as IGPU so llama.cpp does not offload to
  them. Initializing a non-selected device logs a warning.
- Name devices OPENVINO<i> again and show the OpenVINO id in the
  description. Raw "CPU" names shadowed the ggml CPU backend.
- Support GPU.N: create the OpenCL queue on OpenVINO's own context for
  the selected device, and replace "GPU"/"NPU" string comparisons with
  ggml_openvino_is_gpu()/ggml_openvino_is_npu().
- An unavailable GGML_OPENVINO_DEVICE is now an error that lists the
  available devices, instead of silently falling back to CPU.
- Memory: cap iGPU/NPU free memory at system available memory, fall back
  to system memory instead of 0/0 when the plugin lacks memory
  properties, and ignore host USM allocations in GPU usage.
- Initialize the device config once under a lock, even if OpenCL setup
  fails.
- Fix supports_op return type for non-selected devices (build error).

* openvino : take USM entry points from the selected device platform

clGetExtensionFunctionAddressForPlatform was called on the first platform
returned by clGetPlatformIDs. The address it returns is only valid for the
platform it was queried on, and the first platform is not always the one that
holds the device OpenVINO selected.

On a host whose first platform comes from another vendor the lookup returns
null, and then every read, write and memset on a GPU buffer fails with
"clEnqueueMemcpyINTEL not available".

Look both entry points up in init(), on the platform of the device OpenVINO
picked, and keep them in the device config next to the command queue.

Assisted-by: Claude Opus 5

* openvino: fuse MoE experts for models with a fused gate_up weight

FuseMoeCompressed only matches models whose gate and up projections are
separate GatherMatmul ops. gemma-4 packs both into one expert weight and
splits the result after the GEMM, so its MoE block stayed unfused and ran
the expert GEMMs as per-token GEMVs.

Add FuseMoeCompressedFusedGateUp, which matches that shape
(one GatherMatmul -> Slice/Slice -> Gelu(ERF) -> Multiply) and folds it into
the same MOECompressed op, using GEMM3_SWIGLU with GEGLU_ERF. The fused
weight, scale and zero point are split into gate/up halves by copying raw
bytes, since a graph Slice would be rewritten to StridedSlice and constant
folded, whose reference evaluator crashes on sub-byte types.

gemma-4 also applies a per-expert output scale to the down projection before
the router weights. MOECompressed takes only one per-expert weight, so that
scale is folded into the routing weights, which is exact.

The op reads the zero point straight off a weight port and needs an integer
Constant there, so the matcher requires one and leaves natively quantized
experts (exact f16 zp) to the unfused path.

gemma-4-26B-A4B on Arc B390, GGML_OPENVINO_REQUANT_KQUANT=q4_asym64_all,
llama-bench -p 512 -n 128 -r 2, against a GGML_OPENVINO_MOE_OP=0 baseline:
pp512 66.16 -> 1608.73 t/s, tg128 25.94 -> 26.46 t/s. Perplexity over 12
chunks is unchanged (1451.3 +/- 177.9 unfused vs 1427.6 +/- 175.1 fused).

No effect without that requant option, on models with separate gate/up
weights, or on CPU. test-backend-ops -b OPENVINO0 is unchanged by this
commit: two MUL_MAT_ID m_v cases fail, the same two on the unmodified base.

* openvino: fix rank-3 axis handling so MoE works under stateful execution

Stateful execution drops the leading size-1 batch dim, so OV tensors are rank
3 while GgmlOvDecoder::get_shape/get_stride still report GGML_MAX_DIMS=4
reversed entries. Several MoE ops derive OV axis indices straight from that
metadata, so they picked the wrong axis. A MoE model with
GGML_OPENVINO_STATEFUL_EXECUTION=1 aborts while building the graph:

  Check 'is_axis_valid(axis, r)' failed at src/core/src/validation_util.cpp:336
  While validating node 'opset11::TopK ... _ffn_moe_probs ...'
  Axis 3 out of the tensor rank range [-3, 2].

Fix idiom throughout: take the axis from the real OV rank, or shift a
metadata-derived axis down by metadata_rank - actual_rank.

  argsort.cpp    the router top-k axis is 2 on rank 3, not 3. This is the
                 abort quoted above.
  add.cpp        the MoE expert-sum bypass collapses the 8-ADD chain into one
                 ReduceSum on hardcoded axis 2, which on rank 3 reduces n_embd
                 instead of the expert axis. Now rank-2, with the following
                 Unsqueeze at rank-3.
  get_rows.cpp   squeezing a hardcoded {0,1} also strips the batch dim
                 whenever it is 1, which is every decode step. Squeeze down to
                 the trailing two dims instead.
  mul_mat_id.cpp pick the reshape dims by actual rank, and skip the trailing
                 Unsqueeze that re-adds the batch dim.
  view.cpp       the expert-plane slice had the Slice axis, dst_ov_axis, the
                 ShapeOf+Gather index and the Reshape target all rank-4.
  utils.cpp      process_view_input_new's "translate_view already resolved
                 this VIEW, skip re-slicing" shortcut required equal ranks. 4
                 vs 3 never matched, so every resolved expert plane got
                 re-sliced. Now compares the common trailing dims. Same axis
                 shift for the Slice in the view-chain walker.

Stateless is unchanged by construction: every edit is gated on the actual
rank, so axis_shift == 0 reproduces the previous code exactly. Checked on
OV-CPU by diffing greedy output against the unmodified base for dense
gemma-4-E2B, granite-1b-a400m and gemma-4-26B-A4B; all identical.

granite-1b-a400m on OV-CPU aborts with the error above before this change;
after it, it generates and is byte-identical to stateless. Dense gemma-4-E2B
is identical stateless vs stateful both before and after. test-backend-ops
-b OPENVINO0 is unchanged: two pre-existing MUL_MAT_ID m_v cases fail, the
same two on the unmodified base.

gemma-4-26B-A4B is a poor correctness vehicle here. On OV it already drifts
into degenerate repetition a few tokens in, in stateless as much as stateful,
and the two modes diverge somewhere inside that degenerate region instead of
matching token for token. Each mode is self-reproducible across runs.

Known limitation: FuseMoeCompressedFusedGateUp does not match the rank-3
graph, so a MoE model run with GGML_OPENVINO_STATEFUL_EXECUTION=1 loses the
prefill fusion while gaining decode. gemma-4-26B-A4B on Arc B390,
GGML_OPENVINO_REQUANT_KQUANT=q4_asym64_all, llama-bench -p 512 -n 128 -r 2:

  unfused (GGML_OPENVINO_MOE_OP=0)  pp512   66.16   tg128  25.94
  fused, stateless (default)        pp512 1608.73   tg128  26.46
  fused, stateful                   pp512   66.18   tg128  29.91

Stateful is opt-in and off by default, and MoE did not run there at all
before this, so nothing that previously worked regresses. Making the pass
match rank 3 is the follow-up.

* OpenVINO Backend: Upgrade graph cache to use node_idx, src_idx, node type

* ggml-openvino : enable more comprehensive conv fusion

* enable conv ops

* Reject kernel size 0 and support IM2COL_3D

* openvino : abort when the GPU remote context cannot be created

init() logged the error and returned, which left the device name a GPU but
remote_context empty. The remote buffer and tensor paths assert only on the
device being a GPU and then dereference that empty optional.

Those paths have no host fallback, and a device that OpenVINO listed should
have a working OpenCL context, so stop instead of continuing. An OpenCL stack
that is broken as a whole is still caught earlier by the device availability
check, which falls back to CPU.

Assisted-by: Claude Opus 5

* openvino : fix build warnings

The single-argument form of the OpenVINO RTTI macros is the intended one, but
their selector macro leaves __VA_ARGS__ empty, which -Wpedantic reports on
every pass and op header. Turn that warning off for this backend only, the
way ggml-cuda and ggml-sycl already do for their own third-party warnings.

Also drop a break and a dead assignment around a GGML_ABORT, which is noreturn.

Assisted-by: Claude Opus 5

* OpenVINO Backend: Support common MTMD ops

* ggml-openvino: give a reshaping view its own ov::Tensor

* ggml-openvino : compute HARDSIGMOID and EXPM1 in f32

HARDSIGMOID used a 1/6 constant in the input type, which is not exact
in bf16, and EXPM1 lost precision for small inputs in f16. Both now
compute in f32 and convert back, except on NPU where the f32 path
gives wrong results.

Fixes the HARDSIGMOID/EXPM1 test-backend-ops failures on GPU.

* ggml-openvino : update device selection and --list-devices

Show the selecting GGML_OPENVINO_DEVICE value and active device in
--list-devices, startup logs, and backend tests.

Clarify OpenVINO selection uses GGML_OPENVINO_DEVICE, not -dev.

* openvino : remove unreachable OpenCL queue checks

A remote buffer exists only on a GPU device, and init() aborts there if the
queue cannot be created, so the queue is never null at these call sites.

Assisted-by: Claude Opus 5

* openvino : update OpenVINO to 2026.4.1 and GPU drivers to 26.35.39758.10

* docs : update OpenVINO validated models and GPU driver version

* ggml-openvino : skip empty views when giving a reshaping view its own tensor

A zero-size view can sit at the end of a GPU USM buffer (Qwen3.5 recurrent cache). Wrapping it as a remote tensor throws "shared USM buffer has smaller size (0)".

Assisted-by: Claude

* ggml-openvino : rebind the cached decoder when llama passes a different graph

llama keeps separate graphs for batches with and without outputs. llama-server splits the prompt into chunks for context checkpoints, so a cached decoder could be reused with a graph built in other memory and bind the previous chunk's input tensors. SWA and recurrent models then lost most of the prompt in llama-cli and llama-server.

Assisted-by: Claude

* docs : update OpenVINO validated models

Smoke test on Lunar Lake (32 GB) with the two fixes above. Re-add the Qwen3.5 and gemma models.

Assisted-by: Claude

---------

Co-authored-by: Yu, Zijun <zijun.yu@intel.com>
Co-authored-by: ynimmaga <ynimmaga@users.noreply.github.com>
Co-authored-by: Łukasz Ślusarczyk <lukasz.slusarczyk@intel.com>
Co-authored-by: haarika-madaka <haarika.madaka@intel.com>
Co-authored-by: Mustafa Cavus <mustafa.cavus@intel.com>
Co-authored-by: Mostafa Faheem <mostafaaafaheem@gmail.com>
2026-10-03 11:59:25 +03:00
Tarek Dakhran cb7934c52c model : Add LFM2.5-Encoder-350M and LFM2.5-Encoder-230M (#29862)
Register `Lfm2BidirectionalForMaskedLM` architecture for LFM2.5-Encoder
models.
2026-10-03 08:44:45 +02:00
PascalandRuben Ortlam 889edf43dd qwen4exp : halve the indexer score memory (#29825)
* qwen4exp : halve the indexer score memory

The indexer scored all heads in one product and rectified a copy of it,
so two [n_pool, n_idx_h, n_tokens] f32 tensors were live at once, the
largest buffers of the graph at long context. Each head now gets its
own product, rectified and summed in place into one [n_pool, n_tokens]
score.

* qwen4exp: let the allocator reuse the indexer score buffers

Address review from CISC: use plain ggml_add and ggml_relu in the
indexer head loop. The graph allocator already runs them in place when
their source has no other consumer, so the _inplace variants are not
needed. The compute buffer and the speed are unchanged.

* cuda: support 4 heads in the lightning indexer

Dispatch 4 heads to the vector kernel, too few for a wmma tile, and
accept them in supports_op. test-backend-ops covers 4 heads.

* metal: take the lightning indexer head count as a function constant

The kernel reads the head count from a function constant and zero fills
the last head tile, so any head count runs and 64 heads is unchanged.

* qwen4exp: compute the indexer score with the lightning indexer

Address review from am17an: the unweighted sum of the rectified head
scores scaled by 1/sqrt(head_dim) is the lightning indexer with every
head weight set to that scale, so the indexer calls
ggml_lightning_indexer on the pooled keys with an f16 pool mask. The
keys are read once for all heads and no per head score is
materialized.

* vulkan: tile the lightning indexer over keys and tokens

A workgroup scores 64 keys against 8 tokens: the keys are staged once
in shared memory, the queries one head at a time, and each invocation
owns one key for two tokens, so no dot product needs a cross invocation
reduction. The subgroup variant and the flat dispatch are gone, the grid
is keys x tokens x streams.

* vectorize vulkan loads and use fp16 dot product

---------

Co-authored-by: Ruben Ortlam <rortlam@redhat.com>
2026-10-03 07:19:00 +02:00
Xuan-Son NguyenandSigbjørn Skjæret 99b95488ca model: add support for clef decision model (text-only) (#29831)
* init support for clef (text only)

* more static graph

* clean up

* nits

* nits 2

* Update gguf-py/gguf/constants.py

Co-authored-by: Sigbjørn Skjæret <sigbjorn.skjaeret@huggingface.co>

---------

Co-authored-by: Sigbjørn Skjæret <sigbjorn.skjaeret@huggingface.co>
2026-10-03 02:50:48 +02:00
Aman Gupta bed0a85660 CUDA: fuse shared experts into MMVQ (#29184)
* CUDA: fuse shared experts into MMVQ

* check if buffer is null

* move stride_col_dst to fusion args
2026-10-02 21:26:27 +03:00
Sigbjørn Skjæret 4ebdf2c74a ci : use t4-medium for cuda jobs (#29842)
[no ci]
2026-10-02 17:31:19 +02:00
1fb7ef3e33 spec : add probabilistic sampling for simple draft and MTP (#27694)
* Make the drafter probabilistic and the target verify by rejection sampling

* Drop stale spec_draft_q before drafting

* Fallback to argmax sampling for grammar-constrained requests and adding flag for enabling probabilistic draft sampling. Default flag value is greedy.

* Support grammar-constrained requests in rejection sampling

* Fix - renormalize distribution after masking

* copy rng on sampler copy and re-accept drafted tokens on replay

* Fix draft sampler sharing the target's rng stream

* Simplify the rejection sampler's inputs and move replay to the server

* Truncate the draft candidates along with the draft

---------

Co-authored-by: praneshgo <227579474+praneshgo@users.noreply.github.com>
Co-authored-by: Pranesh Gonegandla <pgonegandla@nvidia.com>
2026-10-02 17:52:54 +03:00
Yash Raj Pandey 134b2bb756 ggml-cuda : fix cpy transposed path corrupting non-contiguous dst (#27663) 2026-10-02 22:47:30 +08:00
Yash Raj Pandey 2923cf2862 ggml-quants : avoid invalid rounding in qkx3 scale search (#29817)
* ggml-quants : avoid invalid rounding in qkx3 scale search

The imatrix scale search can produce an infinite, NaN, or otherwise out-of-range value when the fitted minimum collapses to the maximum or makes the range extremely small. That value is then passed to nearest_int and can trip its assertion in Debug builds.

Clamp the quantization level to [0, nmax] before rounding so valid in-range values behave the same as before while invalid scale-search results no longer reach nearest_int.

Add regression coverage for degenerate imatrix groups across q2_K, q4_K, q5_K, q4_1, and q5_1.

Fixes #29804.

Assisted-by: Claude Opus 5.5

* tests: print degenerate imatrix quant types
2026-10-02 17:37:35 +03:00
Yash Raj PandeyandGeorgi Gerganov dd4c286f38 ggml-cpu : fix soft_max_back wrong output when dst aliases src1 (#27096)
* ggml-cpu : fix soft_max_back wrong output when dst aliases src1

GGML_OP_SOFT_MAX_BACK is listed in ggml_op_can_inplace, so the graph
allocator may assign dst to alias either src0 (dy) or src1 (y).

The result was built in several steps:

    ggml_vec_cpy_f32  (nc, dx, dy);
    ggml_vec_acc1_f32 (nc, dx, -dot_y_dy);
    ggml_vec_mul_f32  (nc, dx, dx, y);
    ggml_vec_scale_f32(nc, dx, scale);

When dst aliases src1, the first step overwrites y and the third step
then reads the overwritten values, so the output is silently wrong.
Aliasing dst with src0 is unaffected. The CUDA kernel completes its
reduction before writing and is already safe.

Replace the sequence with a single fused loop that reads both sources
before writing, which is correct under either aliasing.

Add a regression test that marks dy as a graph output so the allocator
is forced to alias dst with y, asserts that the alias actually
happened, and compares against values computed on the host.

* cont : remove comment

---------

Co-authored-by: Georgi Gerganov <ggerganov@gmail.com>
2026-10-02 17:37:11 +03:00
Xuan-Son Nguyen 46ca246de9 model: support nimble decision model (#29844) 2026-10-02 14:56:53 +02:00
Georgi Gerganov d8fbd2583a readme : add cmd install commands (#29850) 2026-10-02 15:39:11 +03:00
Ethan Guo 926862e574 metal : add tensor API flash attention kernel for F16 KV (#29570)
* metal : add tensor API flash attention kernel for F16 KV

* cont : add tensor FA kernels for DK=DV=512 and DK=576, DV=512

* cont : support attention sinks, ALiBi and logit softcap in the tensor FA kernel

* cont : add tensor FA kernel for DK=192, DV=128
2026-10-02 13:19:21 +03:00
Xuan-Son Nguyen a4cb4c61fd llama, server: add /v1/systemone API (models: laya, julia-1, lev, openjev, kev) (#29818)
* init conversion

* convert: ok

* model loaded

* add server code

* improve conversion script

* support shared prompt prefix

* add docs, imorove UX a bit

* add vision support

* add openjev tiny model for testing

* add dev docs

* support lev & kev

* clean up

* fix lev noul

* fix py lint

* nits docs

* clarify about not supporting date_facts
2026-10-02 11:56:04 +02:00
Adrien Gallouët 70849ee82c common : remove fs_open_ifstream() by using u8path() (#29841)
Signed-off-by: Adrien Gallouët <angt@huggingface.co>
2026-10-02 11:43:45 +02:00
Adrien Gallouët 8d81559fa7 llama : silence unused-result warnings (#29839)
Signed-off-by: Adrien Gallouët <angt@huggingface.co>
2026-10-02 11:07:06 +02:00
Adrien Gallouët 6805ae35df llama : use GGML_ABORT instead of throw (#29840)
Signed-off-by: Adrien Gallouët <angt@huggingface.co>
2026-10-02 10:44:44 +02:00
lhez a8c9a4e7cc opencl: use sigmoid f16 for bf16 (#29787) 2026-10-02 11:18:51 +03:00
cwriterandcwriter 392ded6546 SYCL: Q8_0 DMMV ESIMD and MMVQ wide load (#29186)
* Adding wide-load mmvq for Q8_0 and esimd dmmv for q8_0

Assisted-by: Codex

* remove guard for q8_0

* remove docs

* Simplify by committing to clean code without fallback

* Add feature flag as requested

Assisted-by: Claude Opus 5

---------

Co-authored-by: cwriter <cwriter@localhost>
2026-10-02 11:14:35 +03:00
Jiwoong Song 9e258a6e0a vulkan: disable large matmul tile on Samsung GPUs with 32KB shared memory (#28531)
Assisted-by: Claude Opus
2026-10-02 11:13:20 +03:00
Titaniumtown b933289545 sycl: large register file for D=512 FA vec kernels (#29062)
* sycl: large register file for D=512 FA vec kernels

* tests: add 512-wide FA heads to the perf sweep
2026-10-02 11:12:48 +03:00
Łukasz Ślusarczyk c328acc91d sycl : do not use slow oneDNN reference matmul and fattn (#28985)
* sycl : do not use slow oneDNN reference matmul and fattn

* sycl : probe oneDNN matmul once at device init

Assisted-by: Claude Opus 5
2026-10-02 11:11:03 +03:00
Georgi Gerganov 4e2713c162 qwen4exp : optimize mask constructions (#29824)
* qwen4exp : optimize mask constructions

* cont : apply the same change for GLM5-next
2026-10-02 11:08:50 +03:00
Georgi Gerganov 631109b34d ggml : add alloc_buffer_n to buffer type interface (#23671)
* ggml : add `alloc_buffer_n` to buffer type interface

Add alloc_buffer_n method to ggml_backend_buffer_type_i
interface, with a public API ggml_backend_buft_alloc_buffer_n.

- Default implementation in ggml-backend.cpp handles multi-buffer
  splitting and tensor allocation via ggml_tallocr
- Meta buffer type provides custom implementation that creates
  per-device sub-contexts and delegates to simple buffer types
- ggml_backend_alloc_ctx_tensors_from_buft now collects tensors
  into a list and delegates to the new API
- Remove temporary ggml_backend_meta_alloc_ctx_tensors_from_buft
- Add NULL alloc_buffer_n to all existing buffer type
  interfaces (cpu, metal, openvino, hexagon, webgpu, zdnn, virtgpu, repack)

Assisted-by: llama.cpp:local pi

* cont : fix `cur_buf_size` init after flushing a buffer

* ggml : add TODO tag for shared buffer split logic

Assisted-by: pi:llama.cpp/DeepSeek-V4-Flash-Vision-Exp

* tests : add alloc_buffer_n coverage

Assisted-by: pi:llama.cpp/DeepSeek-V4-Flash-Vision-Exp

* cont : fix compile warnings

* tests : add descriptions for alloc_buffer_n tests

Assisted-by: pi:llama.cpp/Qwen3.8-27B

* ggml : address review comments on alloc_buffer_n

- restore GGML_LOG_ERROR on buffer alloc / tensor init failure in the
  default impl (name the failing tensor)
- check the malloc result and drop the _impl indirection in
  ggml_backend_alloc_ctx_tensors_from_buft
- remove comments that restate the code
- fix the TAG_ALLOC_SHARED_BUFFER_SPLIT typo

Assisted-by: pi:llama.cpp/Qwen3.8-27B

* ggml : add get_alloc_size_n to buffer type interface

- Add ggml_backend_buft_get_alloc_size_n public API
- Add optional get_alloc_size_n callback to ggml_backend_buffer_type_i
- Share tensor->buffer planning between alloc_buffer_n default and get_alloc_size_n default
- Replace unchecked realloc with std::vector in alloc_buffer_n default
- Make ggml_backend_alloc_ctx_tensors_from_buft_size use the new API
- Add test-alloc coverage for get_alloc_size_n

Assisted-by: pi:llama.cpp/DeepSeek-V4-Flash-Vision-Exp

* cont : report malloc failure
2026-10-02 11:08:08 +03:00
Aaron Teo 254b177307 ci : fix missing zdnn backend check (#29837)
Signed-off-by: Aaron Teo <aaron.teo1@ibm.com>
2026-10-02 09:10:45 +02:00
Ruben Ortlam fb4b2737a8 vulkan: add logging to pipeline compile issues (#29794) 2026-10-02 08:27:56 +02:00
Kasimir Tanner 207bdab950 pyproject : add linux platform marker to uv torch source (#29177)
In tool.uv.sources, torch was unconditionally pinned to the custom
pytorch CPU index, which lacks macOS Darwin wheels and causes uv sync
to fail on macOS. Add the sys_platform == 'linux' marker to match the
existing Poetry dependencies configuration.

Assisted-by: Antigravity

Resolves: https://github.com/ggml-org/llama.cpp/issues/29176
2026-10-02 07:32:13 +02:00
kurquhar 5fc4f3c8c7 hexagon: install rebuilt HTP skels (#29828)
* hexagon: install rebuilt HTP skels

Assisted-by: OpenCode

* hexagon: fix HTP skel catalog dependencies

Assisted-by: OpenCode
2026-10-01 19:07:38 -07:00
Aman Gupta 159c651f57 qwen4exp: fix tests (#29819) 2026-10-02 09:12:25 +08:00
Jhen-Jie HongandMax Krasnyansky a868c3e3c5 hexagon: add q2_k and q3_k quant type support (#29717)
* hexagon: add q2_k and q3_k quant type support

* hex-qk: consistent allocation of src1_row_size

---------

Co-authored-by: Max Krasnyansky <maxk@qti.qualcomm.com>
2026-10-01 14:12:25 -07:00
Johannes Gäßler ec7630a640 CUDA: fix 2 broken Volta FA cases (#29803) 2026-10-01 21:55:41 +02:00
Johannes Gäßler 78e2964c23 llama: refer to segment documentation [no ci] (#29074) 2026-10-01 21:52:40 +02:00
Adrien Gallouët f1cee9941b common,rpc : fix cache dir creation through symlinks on buggy libstdc++ (#29816)
See https://gcc.gnu.org/bugzilla/show_bug.cgi?id=101510

Close #29759

Signed-off-by: Adrien Gallouët <angt@huggingface.co>
2026-10-01 20:15:02 +02:00
Xuan-Son Nguyen 68e79bd8cd skill: note about model-specific CLI arguments + testings (#29808)
* skill: note about adding model-specific CLI arguments

* add testing instructions
2026-10-01 19:53:23 +02:00
Pascal e358d59178 ci: fix Fusion / metal by updating the qwen4exp baseline (#29812) 2026-10-01 19:57:34 +03:00
Georgi Gerganov 81e39ad343 llama : clamp kpool re-pool bound to existing pools (#29805)
* tests : simplify function signature

* llama : clamp kpool re-pool bound to existing pools

The n_tokens/kpool + n_seqs_unq bound on n_new_g overshoots when a batch
fills the whole cache: n_ctx tokens complete exactly n_ctx/kpool pools, so
the +1 pads new_pool_idxs/new_pool_rep one entry past n_pool_real. Graph
reserve only covers n_pool_real entries, so the first full-context decode
builds bigger tensors than reserved and ggml-alloc demands a graph
reallocation (abort under GGML_SCHED_DEBUG_REALLOC=1).

Clamp the bound to n_pool_real: a ubatch can never mark more pools than
the cache holds, and reserve's n_pool_max already covers that.

Assisted-by: pi:llama.cpp/MiMo-V2.6-Flash-MOPD

* cont : cap to n_pool_max
2026-10-01 19:55:34 +03:00
Yiwei ShaoandMax Krasnyansky dcd387a412 hexagon: shared strided DMA copy for CPY and CONCAT, any-dim CONCAT via DMA (#29685)
* hexagon: shared strided DMA copy for CPY and CONCAT, any-dim CONCAT via DMA

* hex-cpy: various fixes on top of the concat optimizations

Removed CONCAT_DMA_MIN_ROW logic, it was broken with 64-bit DMA.
While it's kinda silly to use DMA for tiny stuff if that tensor gets mapped to an extended buffer the only way to read it is DMA.

Added missing dma_queue_flush() calls.

Added additional guards for conditions we don't support.

---------

Co-authored-by: Max Krasnyansky <maxk@qti.qualcomm.com>
2026-10-01 08:38:37 -07:00
Sam Malayek d775ebf363 server: return HTTP 400 for invalid embedding requests (#29060) 2026-10-01 16:52:43 +02:00
Yu Chengye 2b36825cbc convert : write Gemma embedding scale for DFlash drafts (#29802)
* convert : write Gemma embedding scale for DFlash drafts

A DFlash draft shares the target's token embeddings. Gemma scales them by sqrt(hidden_size) in the forward pass, and the draft config does not state that scale, so the converted draft read unscaled embeddings.

Take the scale from the target config when the draft config has none.

Assisted-by: Claude

* convert : check with get_model_architecture for gemma models
2026-10-01 16:34:32 +02:00
42d958167a cuda : route sm70 to the Turing MMVQ nwarps table (#29753)
* cuda : route sm70 to the Turing MMVQ nwarps table

Volta (sm_70) has no MMVQ parameter table of its own and falls through
to GENERIC, which launches K-quant batch-1 decode (ncols_dst == 1) at
nwarps=4. sm_70 shares TURING's tuning: the K-quant vec_dot prefers
nwarps=2 there. Route sm_70 to the existing MMVQ_PARAMETERS_TURING
table in both the device and the host table selector.

Measured on one Tesla V100 32GB PCIe (PG500-216, driver 580.178.04,
CUDA 12.0.140) with Qwen3.8-27B Q4_K_M, tg128, interleaved A/B in 6
ABBA blocks with paired per-block deltas: +1.091 t/s = +3.17 %
(t = +49.0, all six per-block deltas positive); perplexity
bit-identical (6.3697 +/- 0.04066 both builds, wiki.test.raw). The
patched build's K-quant mul_mat_vec_q kernels launch at nwarps=2
(cubin EIATTR_MAX_THREADS) while Q4_0/Q8_0 stay at nwarps=4, and the
same measurement on the September master base gave +3.84 % (t = 85).

The tuning originates from the V100-focused fork anyei/llamacpp-v100
(MIT), commit b912d1b1e, which carries a dedicated
MMVQ_PARAMETERS_VOLTA table; a cubin-level comparison confirmed that
routing sm_70 to the existing TURING table is equivalent for the
K-quant batch-1 path this change affects, so this is the minimal
2-line form. https://github.com/anyei/llamacpp-v100/commit/b912d1b1e

Original-patch-by: anyei <angelyoelroblesmercedes@gmail.com>

* Update ggml/src/ggml-cuda/mmvq.cu

---------

Co-authored-by: tkittich <tkittich@gmail.com>
Co-authored-by: Johannes Gäßler <johannesg@5d6.de>
2026-10-01 15:21:26 +02:00
Mike van LammerenandNiklas Wenzel 13b4d7135a metal : release temporary private transfer buffers (#29777)
* metal : release temporary private transfer buffers

Assisted-by: OpenAI Codex

* metal : fix order and formatting

---------

Co-authored-by: Niklas Wenzel <dev@nikwen.de>
2026-10-01 16:17:23 +03:00
Masashi Yoshimura 4b1622afb7 webgpu: add bfloat16 support for MUL_MAT/MUL_MAT_ID/GET_ROWS- #29358 (#29358) 2026-10-01 22:09:07 +09:00
Georgi Gerganov 869034b4bb llama : fix invalid assert in recurrent memory (#29799) 2026-10-01 14:49:03 +03:00
ynankaniandJohannes Gäßler b56f34ab13 CUDA: Handle compute type for NVFP4 on cublass path (#29173)
* CUDA: Handle compute type for NVFP4 on cublass path

Signed-off-by: ynankani <ynankani@nvidia.com>

* Use BF16 compute type for quantized models if HW allows

Signed-off-by: ynankani <ynankani@nvidia.com>

* Set acc prec to bf16 for nvfp4 as it needs atleast bf16 range

Signed-off-by: ynankani <ynankani@nvidia.com>

* Update ggml/src/ggml-cuda/ggml-cuda.cu

Co-authored-by: Johannes Gäßler <johannesg@5d6.de>

* preserve op_params for per-expert matmul

Signed-off-by: ynankani <ynankani@nvidia.com>

---------

Signed-off-by: ynankani <ynankani@nvidia.com>
Co-authored-by: Johannes Gäßler <johannesg@5d6.de>
2026-10-01 16:53:52 +05:30
Aman GuptaandGeorgi Gerganov c061df1983 Qwen4Exp: add MTP (#29761)
* Qwen4Exp: add MTP

* remove has_state member, check via ctx_bufs being non-empty

* consistent naming + less verbose comments

* cont : clean-up recurrent memory

* cont : clean-up comments

---------

Co-authored-by: Georgi Gerganov <ggerganov@gmail.com>
2026-10-01 14:13:27 +03:00
Aman Gupta 66e0c17ee1 llama: fix qwen4exp (#29751)
* llama: fix qwen4exp

* qwen4exp: keep kq_mask input the same shape
2026-10-01 14:13:27 +03:00