Commit Graph
1179 Commits
Author SHA1 Message Date
Georgi Gerganov 88aaf325a7 tests : catch vanished rows in the fusion baseline (#30618)
* tests : update the Metal fusion baseline

Re-record the MTL fusion baseline with the current build: 276 rows over 131
dummy models.

Assisted-by: pi:llama.cpp/ggml-org/Qwen3.8-Flash-Next-GGUF (MacBook, M5 Max, 128GB)

* tests : fail test-fusion when a baseline row stops matching

The report was built from the counters the run produced, and the baseline was
only ever used as a lookup table, so a fusion that stopped matching produced no
row and was never compared. qwen4exp and mellum lost their fusions this way
while the Fusion / metal job stayed green.

Report a baseline row that the run does not reproduce as a regression. Scope
the pass to the arches that were actually run so --model keeps working.

Assisted-by: pi:llama.cpp/ggml-org/Qwen3.8-Flash-Next-GGUF (MacBook, M5 Max, 128GB)

* pi : never write inline comments unless the user asks

Assisted-by: pi:llama.cpp/ggml-org/Qwen3.8-Flash-Next-GGUF (MacBook, M5 Max, 128GB)

* tests : keep the mellum dummy model softmax-gated

mellum reads expert_gating_func from the GGUF since #30578, while the fixture
writes sigmoid for every MoE arch, so the Mellum dummy model stopped building
the softmax MoE routing fusion. Write softmax for mellum, matching the arch
default and the GGUFs that conversion/mellum.py produces.

Re-record the MTL baseline: the routing row is back, 277 ok / 0 failed.

Assisted-by: pi:llama.cpp/ggml-org/Qwen3.8-Flash-Next-GGUF (MacBook, M5 Max, 128GB)

* pi : use emoji in the AI stats disclosure

Assisted-by: pi:llama.cpp/ggml-org/Qwen3.8-Flash-Next-GGUF (MacBook, M5 Max, 128GB)

* pi : list the disclosure model after the hardware

Assisted-by: pi:llama.cpp/ggml-org/Qwen3.8-Flash-Next-GGUF (MacBook, M5 Max, 128GB)

* pi : use monochrome symbols in the AI stats disclosure

Assisted-by: pi:llama.cpp/ggml-org/Qwen3.8-Flash-Next-GGUF (MacBook, M5 Max, 128GB)

* tests : add muse-spark to the Metal fusion baseline

Assisted-by: pi:llama.cpp/ggml-org/Qwen3.8-Flash-Next-GGUF (MacBook, M5 Max, 128GB)
2026-10-16 09:35:19 +03:00
Karthik VijayanandKarthik Vijayan f12aaa344c Fix failing RocM tests (#30611)
Co-authored-by: Karthik Vijayan <kavijay@microsoft.com>
2026-10-16 12:46:55 +08:00
Max Krasnyansky b797b9ca81 hexagon: tensor repacking rework and support for tiled and flat quantized and fp weights (#30635)
* hex-repack: Remove need_repack hack and streamline the rules -- breaks GET_ROWS tests for K-Quants, will fix in the next round

Add support for GET_ROWS with tiled Q5_K
Add support for raw k-quant tensors

* hex-get-rows: vectorize raw k-quant ukernels

* hex-binary: refactor dma-loop thread functions (no need for macros, etc)

* hex-quant: move/cleanup flat and tiled dequant ukernels

* hex-repack: add support for repacking/tiling F16/F32/BF16 2D weights

* hex-get-rows: remove redundant cast

* hex-repack: proper filtering for BF16 that were leaking through during the tests

* hex-repack: streamling tiling rules (broken MM for now)

* hex-repack: fix mmid after tiling updates

* hex-repack: rename repacked to tiled in all flags/funcs/kernels/vars

* hex-mm: fix valid_rows calc in various kernels

* hex-mm: remove unused hvx_mm_nx_2d_flat

* hex-mm: cleanup and optimize hvx-tiled vec_dot_f16

* hex-mm: another pass at hvx-tiled f16 matmul

* hex-get-rows: optimize small src1

* hex-get-rows: one more pass at recovering get-rows perf regression

* hex-repack: export GGML_HEXAGON_RT_REPACK via run.py

* hex-mm-tiled: fix accuracy regression with hmx-tiled-f16 matmul

* hex-mm: fold tiled_f16 checks into the general flow

* hex-get-rows: remove dead checks for buffer_is_hexagon and tiling during supports_op

* hex-quant: proper tail handling in new dequant ukernels

* hex-get-rows: remove unused code and cleanup support checks

* hex-mm: cleanup ne1 padding/tiling checks

* hex-mm: cleanup fusion checks for tiled tensors

* hex-repack: add a little can_repack helper to reduce duplication

* hex-repack: never repack host buffers

* hex-repack: factor out repack functions into separate .h and .cpp

* hex-tensor: cleanup tensor ifaces for tiled/padded sizes

* hex-repack: update k-quant tiling comments and cleanup #defines

* hex-dcache: factor out l2/dcache dirty tracking and flushing

* hex-tensor: update docs and remove #defs duplication

* hex-tensor: remove unused/incorrect get_data interface

* hex-get-rows: cleanup vtcm layout build

* tests: disable huge MM tests

* hex-dcache: add missing header
2026-10-15 20:30:11 -07:00
Said Taghadouini 2f729cfa4c mtmd: lightonocr does not require longest edge to always be 1540, native resolution is the default (#30600)
Assisted-by: Codex
2026-10-15 23:53:02 +02:00
ba3c05f71d model: add Muse Spark Open (text, vision and audio) (#30616)
* model: add Muse Spark Open (text, vision and audio)

Muse Spark Open is a latent space MoE: the residual stream stays at 8192
while the 256 experts live in 4096, with a projection down and up around
them, top-8 sigmoid routing with a correction bias. Attention interleaves
three sliding layers at a 2048 window with rope for one global layer
without rope, with a sigmoid output gate and a weightless QK-norm.
Residuals are norm preserving, their gate is folded at conversion time.

Vision reuses the Muse Glimmer tower. Audio is a causal convolutional
pyramid with no attention and no spectrogram, four stages folding 16, 6,
5 and 4 samples into the channel dim, one embedding per 80 ms of 24 kHz.

The quantizer keeps the latent projections and the attention gate at
Q8_0, a little over 1% of the weights, since the whole residual stream
passes through them and the low-bit types turned the output into a
repeating loop. The model saver writes the post norm epsilon, and the
arch is registered in test-llama-archs.

Co-authored-by: Xuan Son Nguyen <son@huggingface.co>
Co-authored-by: Pedro Cuenca <pedro@huggingface.co>

* Muse Spark: drop the redundant forward expand

* Muse Spark: drop the redundant forward expand and tighten the comments

* Update src/llama-quant.cpp

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

* Update src/llama-graph.cpp

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

* Update conversion/muse_spark.py

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

* Update conversion/muse_spark.py

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

* Update conversion/muse_spark.py

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

* Update conversion/muse_spark.py

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

---------

Co-authored-by: Xuan Son Nguyen <son@huggingface.co>
Co-authored-by: Pedro Cuenca <pedro@huggingface.co>
Co-authored-by: Sigbjørn Skjæret <sigbjorn.skjaeret@huggingface.co>
2026-10-15 23:27:12 +02:00
Karthik VijayanandKarthik Vijayan 3e4dfa8f39 models : add maion-coder (#29778)
* Add Support to maion-coder arch

* iAdd Support to support Draft models

* Address comments

* Remove maion context shift function

---------

Co-authored-by: Karthik Vijayan <kavijay@microsoft.com>
2026-10-15 12:18:46 +03:00
Yuri Khrustalev 82e1e2fd84 jinja : write nan and inf in tojson as json.dumps does (#30483)
tojson wrote nan, inf and -inf for these floats, as values and as dict
keys; json.dumps writes NaN, Infinity and -Infinity. The string and safe
filters of a float keep nan and inf, as Python's str() does, so they no
longer use the tojson function. float_to_json is renamed float_repr, as
it now gives str(); float_json gives the json.dumps text.

Also say in a comment why a last digit 9 is not bumped at a power of 2.
2026-10-15 10:49:40 +02:00
losewayy 47f5a64ae1 vulkan : fix concurrent first device init (#29766)
* vulkan : fix concurrent first device init

* tests : load models in parallel in test-thread-safety

Loading them serially never exercised concurrent first backend init, so the crash reported in #29764 was invisible to this test.

* vulkan : create fence before retaining device in buffer context
2026-10-15 10:37:35 +02:00
William Tan 2130ca8476 jinja : fix loop body variable scoping (#30533)
Assisted-by: Codex
2026-10-15 09:14:01 +02:00
Simone IacominoandSigbjørn Skjæret 2ce20d22fd unicode : reject overlong UTF-8 and out-of-range codepoints (#30485)
* unicode : reject overlong UTF-8 and out-of-range codepoints

* Update tests/test-unicode.cpp

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

* tests: document byte-wise UTF-8 error recovery

* break out of first test on failure

---------

Co-authored-by: Sigbjørn Skjæret <sigbjorn.skjaeret@huggingface.co>
2026-10-15 10:12:56 +03:00
Nathan WilsonandRuben Ortlam 733f57a999 vulkan : copy strided f16 KV once on AMD so reads spread across memory channels (#27703)
* vulkan : copy strided f16 KV once so reads spread across memory channels

Assisted-by: Claude (Opus 4.8 / Opus 5)

* tests : cover the f16 KV copy to scratch

Assisted-by: Claude (Opus 4.8 / Opus 5)

* vulkan : gate the f16 KV scratch copy to AMD

Assisted-by: Claude (Opus 4.8 / Opus 5)

* vulkan : reuse the generic copy shader for the f16 KV scratch fill

The f16 clause needs only a strided copy, not a dequant, so drop the
dedicated shader and dispatch the existing copy pipeline instead. The
copy is described in source memory order so the strided access lands on
the writes, which stream, instead of the reads, which would concentrate
on a few memory channels.

Assisted-by: Claude (Opus 4.8 / Opus 5 / Fable 5)

* vulkan : extend the strided KV scratch copy to bf16

Assisted-by: Claude Opus 5

* vulkan : clean up comments

Assisted-by: Claude Opus 5

* consolidate copy shaders

* disable on RDNA4

---------

Co-authored-by: Ruben Ortlam <rortlam@redhat.com>
2026-10-15 10:09:02 +03:00
cqderekandravi9 5a638aedb7 cpu: add F16 SWIGLU_OAI reference implementation and validation (#29200)
* cpu: add F16 SWIGLU_OAI reference implementation and F16 activation op test coverage

Adds ggml_compute_forward_swiglu_oai_f16() to the CPU backend, mirroring
the existing F32 path, and extends test-backend-ops with F16 test cases
for SWIGLU_OAI so downstream backends(e.g. Hexagon) can be CPU-diffed
for correctness.

* ggml-openvino: fix f16 SWIGLU_OAI type mismatch in alpha/one constants

translate_glu_swiglu_oai creates alpha_const and one as f32 constants
unconditionally, so f16 inputs hit a type mismatch in Swish/Add on GPU
("Swish inputs must have the same type but they are: f16 and f32").
translate_glu_swiglu_clamp already upcasts to f32 and converts back at
the end; apply the same pattern here.

---------

Co-authored-by: ravi9 <ravi.panchumarthy@intel.com>
2026-10-15 10:07:45 +03:00
Ethan Guo 6e66cb5d4e metal : add BF16 variant of the tensor API flash attention kernel (#30491)
* metal : add BF16 variant of the tensor API flash attention kernel

* cont : move tensor FA kernels to kernels/fa_tensor.metal
2026-10-15 09:39:47 +03:00
PascalandGeorgi Gerganov e467e5b5e5 kv-cache: honor PARTIAL_ONLY in the plain KV cache (#30514)
* llama: honor PARTIAL_ONLY in the plain KV cache

A non-SWA llama_kv_cache rolls back with seq_rm, so it holds no
partial state, yet state_write ignored the flag and serialized every
cell. A draft context made of a plain KV cache (MTP head, dense -md
draft) therefore copied its whole KV into each server checkpoint,
growing it by the draft KV size per token.

The cache now writes and reads nothing for PARTIAL_ONLY unless it is
SWA, and the iSWA wrapper no longer gates its base cache since it
filters itself.

* Update src/llama-kv-cache.cpp

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

* llama: delegate PARTIAL_ONLY to the caches of every memory wrapper

The hybrid, hybrid-idx and dsa-iswa wrappers skipped their attention
caches on PARTIAL_ONLY, which also dropped an SWA window that is
partial state. Each cache now applies the flag itself, so the
wrappers only delegate. The indexer is marked stale only when a
restore rewrites cells.

test-llama-context gains a case checking that the partial state is
not larger than the full state and, without SWA, does not grow with
the tokens.

---------

Co-authored-by: Georgi Gerganov <ggerganov@gmail.com>
2026-10-15 07:39:18 +03:00
62df1815bc cuda, hip: GDN chunked kernel (#29353)
* cuda, hip: GDN chunked kernel

Ported-from: https://github.com/tetherto/qvac-fabric-llm.cpp/pull/257
Source-commit: 53f0027d8070715fe9ad2131f3201bb0001d4eed
Co-authored-by: Aman Gupta <aman.gupta@tether.io>

* fix older builds

* review fixes

* cuda: enable GDN MMA on tested Ampere and Ada GPUs

Enable SM86 and SM89 after RTX 3090 and RTX 4090 correctness and prefill benchmarks. Keep other untested architectures on the existing fallback.

Assisted-by: Codex

* cuda : use refactored swizzle API in GDN MMA

* ggml : GDN only guarantees the final state when n_tokens > K

* cuda : add short GDN MMA phase comments

---------

Co-authored-by: gianni-cor <gianfranco.cordella@tether.io>
Co-authored-by: Aman Gupta <aman.gupta@tether.io>
2026-10-14 22:00:58 +08:00
Pascal 6868a52cb3 ggml: fix conv_1d result layout for batched input (#28738)
* ggml: fix conv_1d result layout for batched input

The GEMM behind conv_1d folds the batch into its rows, so the product
sits in memory as [OL, N, OC] while the reshape claims [OL, OC, N].
The shape is right and the contents are transposed, so nothing fires:
no assert, no abort, the graph just runs and the channels come out
interleaved with the batches. With N == 1 the two readings describe
the same buffer, which is why the three vocoders in the tree never
hit it, they all squeeze to 2d right after the call.

Reshape to the real layout and permute OC and N back. The N == 1 path
keeps its direct reshape, so the current callers build the same graph
and pay for no extra copy.

tests/test-conv-1d.cpp compares one batched convolution against N
single batch convolutions built from the same weight and the same
input slices. Five of the seven geometries fail before this change and
all seven pass after. N == 1 and OC == 1 pass either way, since
swapping two dimensions when one of them is 1 changes nothing.

The batched path matters to downstream audio projects where several
synthesis lanes share a single graph. Nothing in mtmd batches a
vocoder today, each one streams one flow and carries its left context
between calls, but the op is correct now if one of them grows a lane
dimension.

* ggml: build the same conv_1d graph for every batch size

Address review from ggerganov: drop the N == 1 branch of ggml_conv_1d.
The GEMM result is always reshaped to [OC, N, OL] and permuted back,
as ggml_conv_2d does, so every batch size builds the same graph. For
N == 1 this adds a cont on an already ordered tensor.
2026-10-14 14:16:08 +03:00
Georgi Gerganov 3fb47677ea common : add --spec-draft-n-ubatch (#30459)
* common : add --spec-draft-n-ubatch

Allow configuring the physical batch size of the draft context. The
drafters feed it at most the drafted tokens per sequence and chunk larger
batches by its own n_ubatch, so it does not need the ubatch of the target.
A smaller value shrinks the draft compute buffer and leaves more VRAM for
the target KV cache (the draft is fitted together with the target).

The default 0 keeps the current behavior of following --n-ubatch.

Assisted-by: pi:llama.cpp/Qwen3.8-Flash-Next

* cont : remove comment

[no ci]
2026-10-14 13:58:05 +03:00
Georgi Gerganov 8c2134a875 ggml : add note about mul_mat_id + disable mul_mat_id_dup tests (#30501) 2026-10-14 12:36:52 +03:00
Yuri Khrustalev 77390baa18 jinja : write floats in tojson as Python does (#30362)
tojson printed floats through a C++ stream (6 significant digits, no
".0"), so 48.858844 became 48.8588 and 1.0 became 1 in tool definitions,
tool-call arguments and other template output. Write the shortest digits
that read back as the same value, as json.dumps does. The string and safe
filters of a float use the same code.
2026-10-14 09:18:38 +03:00
Sandro Hanea 10da8c463f unicode : add opt-in NFC normalization for Unigram tokenizers (#30390)
* unicode : add opt-in NFC normalization for Unigram tokenizers

Note: used help from Codex, Sol 6.1 Medium

* Fixing initial bot comments + linter error (using existing out instead of print)

* removed test_nfc_tokenizer
2026-10-14 08:33:24 +03:00
Yuri Khrustalev 04f7226a2c server : escape the special tokens without std::regex (#30243)
The d1-omni and kev prompts escape "<|name|>" in the request text with
std::regex_replace. libstdc++ recurses for each char of the name, so a
state of "<|" and 30000 letters overflows the stack of the HTTP thread
and llama-server dies with SIGSEGV.

server_decision_escape_special_tokens() writes the same text with a loop.
test-server-component compares it with the regex on every string of up to
6 chars of "<|>a-".
2026-10-14 01:17:11 +02:00
Xuan-Son Nguyen ff026f8c0a chat: replace output string with parts (#30389)
* chat: replace output string with parts

* reserve first
2026-10-14 01:03:56 +02:00
Jeff Bolz 77c4540ce4 vulkan: Optimize pool_2d when the pooling window is large (#30119)
When the window is larger than a subgroup, use a whole subgroup to
compute each output.
2026-10-13 17:17:59 +02:00
Georgi Gerganov a17e94d229 tests : consolidate context state tests into test-llama-context (#30023)
* tests : consolidate the context state tests into test-llama-context

test-save-load-state, test-recurrent-state-rollback and
test-state-restore-fragmented check the same object from three places:
the KV cache and the recurrent state of a llama_context, saved, loaded,
copied, rolled back. Each had its own main, its own loop over the
models, its own copy of the token and the comparison helpers, and a run
of all three loaded the same models several times.

They are one binary now, and the case is the unit of the test:

  tests/llama-context/test-case.h   the interface a case implements
  tests/llama-context/cases-kv.cpp  the cases of the KV cache
  tests/llama-context/cases-rs.cpp  the cases of the recurrent state
  tests/llama-context/test-util.cpp tokens, generation, comparison
  tests/llama-context/test-runner.cpp the models of a case, on threads
  tests/llama-context/test-log.cpp  the log file of a model

A run is a table: a row per model, a column per case, a cell holds the
verdict of that pair. The models load on several threads (-j, --threads,
min(cores, 4) by default), the table is printed when it is complete and
the log of a model that failed is printed after it, in the order of the
rows. A case returns a verdict and the reason for it, and the runner is
the only one who reports - the test does not use tests/testing.h, a run
over 130 models is a matrix of verdicts and not a tree of subtests.

Staging no longer travels through the environment: llama_model_set_attn_rot
replaces the LLAMA_ATTN_ROT_DISABLE variable, which a model loaded in one
thread used to change for the models the other threads were loading, and
llama_memory_get_buffers is added for the cases that copy the buffers of
a memory. common_log_set_stdo keeps the console quiet while the table is
being built.

ggml_aligned_malloc allocates with posix_memalign when the thread
sanitizer is on. On macOS it allocates with the Mach VM, which the
sanitizer does not intercept, so it never sees a page range change hands
and reports the first write into recycled memory as a race - 30 of them
in this test, and every one of them false.

Assisted-by: pi:llama.cpp/Qwen3.8-Flash-Next

* cont : run single-threaded by default

* cont : try to build the tests on all platforms

* cont : trim down some body fat

* cont : fix

* cont : fix the fix
2026-10-13 16:21:55 +03:00
losewayy 411d548515 vulkan : split matmul dispatch when it would exceed maxComputeWorkGroupCount (#29533)
* vulkan : split matmul dispatch when it would exceed maxComputeWorkGroupCount

* vulkan : fix narrowing in chunked matmul dispatch

* vulkan : zero fill the A side of mul_mmq for k steps past the end

The A side skipped the shared memory write for the k steps past end_k, while the B side was already told the bound and zero filled. The A registers are read for all BK_STEP steps, so the stale scale went into the dot product and produced NaN at k = 32.

Pass the bound to block_a_to_shmem and write zeros when it is false, for every DATA_A type, mirroring the B side.

Also assert that the B and D subbuffer sizes are not VK_WHOLE_SIZE before they are adjusted per chunk, and add two batched cases of the chunked shape.

Signed-off-by: losewayy <104015127+losewayy@users.noreply.github.com>

* vulkan : remove redundant matmul tile adjustment

---------

Signed-off-by: losewayy <104015127+losewayy@users.noreply.github.com>
2026-10-13 13:03:47 +02:00
kurquharandMax Krasnyansky 4413a19241 hexagon: extend scalar DMA broadcasting support for binary ops (#30245)
* tests: cover weighted scalar broadcasts

Assisted-by: OpenCode

* ggml : support extended scalar DMA binary ops

Assisted-by: OpenCode

* hex-binary: remove all non-dma paths

* hex-binary: add support for non-contig strides and cleanup dma preable a bit

* hex-binary: cleanup the gating and simplify dma retry

* hex-geglu-erf: add a gate for F32 only, we dont support F16 for this op

* tests: cover weighted repeat broadcast

Assisted-by: OpenCode

* hex-binary: update REPEAT and COMPLEX kernels to use dma as well (missed them last time)

* hex-add-id: add dma support for src2 (idx)

---------

Co-authored-by: Max Krasnyansky <maxk@qti.qualcomm.com>
2026-10-12 20:33:05 -07:00
Xuan-Son Nguyen 097cf33d94 model: support vllm decision 2 model (#30378)
* models: support pplx-decider

* vllm d2

* update

* clarify naming rule

* stricter validation

* fix ci

* rm redundant filter_tensors
2026-10-12 13:39:55 +02:00
Georgi Gerganov ccdbe29053 tests : link each test against only the libraries it uses (#30299)
Give llama_build() and llama_build_and_test() a LINK keyword (default:
llama-common) so call sites can declare what a test actually needs:

- ggml-only tests now link just ggml: test-opt, test-metal-graph-optimize,
  test-backend-ops, test-tiled-mulmat, test-barrier, test-quantize-fns,
  test-quantize-perf, test-rope, test-col2im-1d, test-gguf, test-alloc
- llama-only tests drop llama-common: test-unicode, test-sampling,
  test-grammar-parser, test-llama-grammar, test-batch-alloc,
  test-gbnf-validator, test-quant-type-selection
- both mtmd tests link mtmd only, dropping the LLAMA_TEST_NAME /
  target_link_libraries workarounds

Both helpers now set cxx_std_17 explicitly, since the tests no longer
inherit it from llama-common's PUBLIC interface.

tests/testing.h no longer includes common.h; the only symbol it used from
there (string_join) is inlined.

Assisted-by: pi:llama.cpp/Qwen3.8-Flash-Next
2026-10-12 13:32:56 +03:00
Jeff BolzandAnjielon 966baae76b vulkan: handle mul_mat_id duplicates in the prepass rather than looping (#29998)
* vulkan: compute every row of an expert in mul_mm_id when ids repeat

* vulkan: handle mul_mat_id duplicates in the prepass rather than looping

Extend the "hoist row ids" optimization to always be enabled and to emit a
compact list of tile descriptions that need to run, and to emit multiple tiles
when needed. Then we can launch a tighter upper bound on the number of
workgroups and the tail can trivially early exit.

---------

Co-authored-by: Anjielon <[EMAIL_REDACTED]>
2026-10-11 14:02:57 +02:00
Xuan-Son Nguyen 62a6f74bab server: add component tests (#30312)
* server: add component tests

* nits

* add codeowners
2026-10-11 12:51:07 +02:00
Xuan-Son Nguyen 781dbc5ac9 spec: properly handle mtmd input for mtp (#30257)
* spec: properly handle mtmd input for mtp

* nits
2026-10-10 11:22:38 +02:00
Anav Prasad 50e3e3e480 CUDA: Remove redundant CUDA copies after SSM_SCAN (#29807)
* CUDA: fuse copy of updated state snapshots into recurrent cache with ssm_scan

* CUDA: remove redundant cuda copies with K==1 (non spec-dec) scenario as well
2026-10-09 18:23:28 +02:00
Aldehir Rojas 8b54361025 chat : refactor API (#30210) 2026-10-09 10:06:22 -05:00
Mark TodorovichandMax Krasnyansky d4d82d67f4 hexagon: fix IM2COL patch-embed DMA ring overflow (#30189)
* hexagon: fix IM2COL patch-embed DMA ring overflow

The exact-tiling (stride == kernel, no pad/dilation) IM2COL DMA kernel
issues IC*KH DDR->VTCM descriptors per output row without checking the
return value of dma_queue_push(), and then pops IC*KH times. The per-thread
DMA ring holds 256 entries and a push into a full ring returns false
and drops the transfer, so for IC*KH > 255 the remaining rows of the
VTCM staging buffer were never written and stale data (often NaN/inf)
leaked into the output.

Solution is to retire the oldest descriptor when the ring is full, just as the blocked
kernel in the same file already does, and wait with dma_queue_flush().

For testing, added exact-tiling test cases with IC*KH > 256 (2D 1x1, 2D 2x2 patch
embed and 1D, F16 and F32 dst), which fail on HTP without this fix.

* Apply suggestion from @max-krasnyansky

---------

Co-authored-by: Max Krasnyansky <maxk@qti.qualcomm.com>
2026-10-09 08:04:09 +02:00
71ad0590f4 CUDA: improve top-k algorithm selection (#28713)
* CUDA: radix top-k for large row counts

Replaces CUB's per-row DeviceTopKKernel with a grid-over-rows radix select,
gated on GGML_CUDA_TOPK_RADIX_MIN_ROWS. On qwen4exp at 34,816 tokens this cuts
top-k from 1,671,253 launches / 5,761.8 ms to 2,329 / 941.8 ms.

* CUDA: select the TOP_K implementation by shape

Replace the nrows/ncols special case with the decision boundary from #28547
(as implemented in #29278): bitonic for short rows, radix select for several
long rows, and DeviceTopK or CUB argsort for a single long row. The
thresholds stay overridable at build time.

Two refinements on top of that boundary:
- bitonic stays in use for rows up to a padded 1024 while the rows fit in one
  wave of blocks (nrows <= number of SMs); radix select pays a fixed cost of
  about a dozen launches that only amortizes over more rows
- with DeviceTopK available, it handles up to two rows

Radix select now processes rows in chunks so its scratch memory stays bounded,
and the bitonic path keeps its chunking. HIP and MUSA keep their previous
thresholds.

Add perf cases around the bitonic/radix crossover to test-backend-ops.

* CUDA: make top-k comments less verbose

* CUDA: remove the TOP_K width limit from supports_op

* CUDA: use DeviceTopK for single-row TOP_K if available

* CUDA: avoid ncols overflow in the TOP_K bitonic check

* CUDA: share the row chunking helper between argsort and top-k

* CUDA: do the TOP_K radix blocks_per_row math in int64_t

* CUDA: rename GGML_CUDA_TOP_K_NROWS_THRESHOLD_DEVICETOPK to GGML_CUDA_TOP_K_NROWS_THRESHOLD

* CUDA: share one sort helper between the bitonic and CUB TOP_K paths

* CUDA: update the TOP_K TODO, threshold and chunking comments

* tests: add TOP_K cases that span several row chunks

* CUDA: use int64_t col in the TOP_K radix loops, fix threshold comment

* CUDA: limit TOP_K and ARGSORT support to ne[0] <= INT_MAX

---------

Co-authored-by: praneshgo <227579474+praneshgo@users.noreply.github.com>
Co-authored-by: Pranesh Gonegandla <pgonegandla@nvidia.com>
2026-10-08 18:34:27 +02:00
Prabhsimran Singh c35b66744f CUDA : looped PAD kernel for more than 65535 rows or slices (#30147) 2026-10-08 15:36:53 +02:00
Aman Gupta c811cb8f0a llama: support MoE cache over multiple GPUs (#30112) 2026-10-08 15:50:49 +05:30
gianni-cor ff5888f999 vulkan : fix TOP_K for +inf/NaN inputs and k = 1 on negative values (#30107)
The bucket search in topk_nary_search.comp started from the range
[0, 0xFF800000), which ends just below the ordered-uint mapping of +inf,
so +inf and NaN were never counted. A workgroup block with fewer than k
countable values left the ballot empty and the shader read uninitialized
shared state (hang/device lost on NVIDIA, wrong indices on AMD), and a few
+inf in a block were selected without being counted, dropping real top
values.

Map NaN to -inf on input, start from [0, 0xFFFFFFFF) so every value is
counted, and clamp the top bucket's end (2^32) instead of wrapping to 0.

The k = 1 path compared float bits as signed integers, which orders
negative values backwards; compare floats instead.

Add test_top_k_inf to test-backend-ops: negative values, fewer than k
+inf and many -inf, for k = 1, 10, 40.

Assisted-by: Claude Opus 5.5
2026-10-08 12:14:35 +02:00
Jeff Bolz dac3087394 vulkan: extend sparse FA support to coopmat2 (#30003) 2026-10-08 11:36:20 +02:00
bri-prism 46baf1f1fe sycl: FWHT optimizations (#29605) 2026-10-08 02:50:22 -04:00
Eden fda1866613 CUDA: fix norm family kernels when ne[2]/ne[3] exceed grid dim limits (#28175) 2026-10-08 09:46:21 +03:00
Titaniumtown ff30363a0e sycl: fuse the delta-net alpha gate (add + unary + mul) (#29687)
* sycl: fuse the delta-net alpha gate (add + unary + mul)

* tests: cover the fused add + unary + mul chain

* sycl: give the fused alpha gate a flat path and pin the node skip
2026-10-08 09:45:21 +03:00
Nicolas Budyn bd4eeaa047 chat: fix jinja parser for TranslateGemma (#30096)
* chat: fix jinja parser for translategemma

* log if missing soruce_lang_code or target_lang_code
2026-10-07 22:48:06 +02:00
Frost-54andAlde Rojas 5de733437b chat : name tool and argument parser rules by index (#30088)
* bugfix: infinite recursion caused by a tool named 'call'(#29967)

* chat : use index for schema and argument rules

* tests : remove tests

* tests : add expect_rules to peg test parser

---------

Co-authored-by: Alde Rojas <hello@alde.dev>
2026-10-07 22:37:51 +02:00
Aman Gupta d6cf9acb25 llama : add a GPU cache for MoE experts kept in host memory (#29887)
* llama : add a GPU cache for MoE experts kept in host memory

Assisted-by: Claude

* use llama_moe_cache_ptr
2026-10-07 21:07:45 +03:00
ShobhitandGeorgi Gerganov 42c787e8c1 cuda: update uncoalesced memory reads in pool2d (#29425)
* cuda: update uncoalesced memory reads in pool2d

* cuda: Added perf and __restrict__ in pool2d

* Update POOL2D_WARP_KERNEL_MIN_WINDOW macro

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

* cuda: Fix compiler bugs

---------

Co-authored-by: Georgi Gerganov <ggerganov@gmail.com>
2026-10-07 19:53:43 +02:00
Aldehir Rojas 18b5f8b186 server : accumulate generated text and tokens as parse input (#29876)
* server : collect text and token input

* common : add a tokenize helper that aligns tokens with bytes

* common : simplify tokenize logic
2026-10-07 11:49:17 -05:00
pratiknarola-t 988190680d metal : few-row MMA mat-mul for the remaining src0 types (#30065)
The generic few-row MMA kernel works for any type with a 16-weight
dequantizer, so it now also takes BF16, Q1_0, Q2_0, MXFP4, Q2_K, Q3_K,
TQ2_0 and the IQ types. Each type starts at the row count where it beats
the current kernels on an M3 Ultra: 5 rows for TQ2_0, 4 for BF16, 3
for MXFP4, Q2_0, Q2_K and IQ4_NL, and 2 for the others.

test-backend-ops perf -o MUL_MAT, m=4096, k=14336, M3 Ultra, time of this
change over master (mean of two interleaved runs each): 0.23 to 0.98 from
the threshold to 8 rows, 0.24 to 0.33 at 9 to 16 rows, and 0.99 to 1.01 at
1 and 512 rows.
2026-10-07 16:49:37 +03:00
Niko MaroulisandGeorgi Gerganov 7e8324f5fe metal : fix MUL_MAT+ADD fusion when the residual is itself a MUL_MAT (#30100)
* metal : fix MUL_MAT+ADD fusion when the residual is itself a MUL_MAT

ggml_metal_op_mul_mat_mma picks the residual of a fused MUL_MAT+ADD as
"the ADD operand whose op is not MUL_MAT". When both operands of the ADD
are mat-mul outputs (x = W1 @ u + W2 @ v), that test is true for both, so
the residual resolves to the fused mat-mul's own, never-written output and
the kernel adds whatever that buffer holds.

The fusion check (ggml_metal_mul_mat_add_operand) already selects the
operand by identity; make the encoder do the same.

Clef decision models hit this in their head (proj_option_context @ ctx +
proj_option_lexical @ lex, 9 option rows): on Metal, /v1/systemone
probabilities collapse toward uniform (billing 0.28 where the CPU backend
gives 0.977, Cloudflare_clef-flash Q8_0), deterministic per memory layout,
correct with GGML_METAL_FUSION_DISABLE=1. Not a quantization issue: the
same file is right on CPU.

Add a MUL_MAT_ADD mode to test-backend-ops where the residual is a second
mat-mul; on Metal it fails 27 of 28 cases before this change (the one pass
is f16 n=2, under the MMA row threshold, so nothing fuses).

* Update tests/test-backend-ops.cpp

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

---------

Co-authored-by: Georgi Gerganov <ggerganov@gmail.com>
2026-10-07 16:41:50 +03:00
Hrishith ThadicherlaandGeorgi Gerganov d0b490f25e sampling : use greedy selection for eligible temperature-zero chains (#29797)
* sampling : use greedy selection for eligible temperature-zero chains

Assisted-by: OpenAI Codex

* Apply suggestion from @ggerganov

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

* sampling: simplify zero-temperature greedy eligibility

Allow the same greedy selection on CPU and grammar/reasoning-budget paths.
Keep distribution sampling for dynamic temperature and requested probabilities.
Cover the common sampler selection and probability behavior in the existing sampler tests.

Assisted-by: OpenAI Codex

* sampling : use greedy selection after final top-k with k=1

Assisted-by: OpenAI Codex

* cont : clean-up

---------

Co-authored-by: Georgi Gerganov <ggerganov@gmail.com>
2026-10-07 13:34:18 +03:00