Commit Graph
1392 Commits
Author SHA1 Message Date
Giulianno 9309bf1bcf model : Add support for Metis-1.6 models (#31134)
Assisted-by: Claude
2026-11-03 16:24:45 +01:00
Tarek Dakhran d1044bc827 model : enable NVFP4 LFM3 ggufs (#31608) 2026-11-03 15:36:44 +01:00
Pascal 749db7f034 qwen4exp: write the conv state tail without an intermediate cont (#31443)
build_conv_state_at makes a contiguous copy of the conv input tail
before writing it into the recurrent state, so every GDN layer issues
two device copies per generated token where one is enough. On
Qwen3.8-Flash-Next (36 GDN layers) that is 72 copies per token, now 36.
Speculative decoding adds one more per rollback slot. ggml_cpy reads
the strided view directly, as the shared build_conv_state already does,
so the extra copy is dropped.
2026-11-03 13:47:23 +02:00
Tarek Dakhran f9448cddc4 model : record layer inputs in LFM3 for spec dec (#31598) 2026-11-03 13:29:10 +02:00
Aman GuptaandPratik Narola 259d91709d meta: support GLM-5.3 next sm tensor (#30499)
* meta: support GLM-5.3 next sm tensor

* glm5-next : cont q_absorbed for -sm tensor with non-unified KV

* meta : keep GGML_OP_ARANGE in ggml.h order

* glm5-next : split MLA heads in pairs for -sm tensor

The MLA cache has a single KV head, so the heads on a device are its GQA
ratio. The HIP tile FA kernel for D=512 needs an even GQA ratio; with 3
devices the 16 heads of the test model were split 5/5/6 and it aborted.

* meta : keep the split for views inside one slab of the split axis

With a non-unified KV cache, build_attn_mha splits q into streams with a
view. For a permuted q, e.g. the absorbed query of glm5-next, that view
aborted with "view of permuted tensor not implemented". The split is now
kept if every other dim of the source and the view stays inside one slab
of the split axis. The per-device strides are only rescaled above it.

From tetherto/qvac-fabric-llm.cpp#353.

Co-authored-by: Pratik Narola <pratik.narola@tether.io>

* glm5-next : drop the q_absorbed cont

The meta backend can now split the per-stream view of the permuted q, so
the cont is not needed for -sm tensor with a non-unified KV cache. With
the cont, the q of flash attention was not contiguous, and when FA ran on
the CPU (WebGPU) the meta get_tensor aborted.

---------

Co-authored-by: Pratik Narola <pratik.narola@tether.io>
2026-11-03 10:51:10 +02:00
Aman GuptaandPascal b74298c109 delta_net_base: refactor conv state (#30878)
* delta_net: access conv state via views

* fix sm tensor

* delta_net: fill the conv slot index on the host

The rollback slot index of the conv windows is a constant per graph
shape. It is now an I32 input of the recurrent memory filled on the
host, instead of an arange and an argsort computed on the device, and
the meta backend no longer needs a mirrored split for ARANGE.

---------

Co-authored-by: Pascal <admin@serveurperso.com>
2026-11-03 08:28:35 +01:00
Georgi Gerganov ca8daad199 llama : drop the redundant cnv from llm_attn_mask (#31558)
The dspark block mask is now derived at input-fill time: llm_attn_mask::block_uniform
fills every token of a sequence with the mask row of its first token, so the
graph-level reshape/repeat in dflash is gone and cnv always aliases raw.

Assisted-by: dsh:llama.cpp/ggml-org/GLM-5.3-Flash-GGUF (Mac Studio, M5 Ultra, 256GB)
2026-11-03 08:11:06 +02:00
Aman Gupta 0f663d2471 dsv4.1: fix llama-context tests with partial offload (#31292) 2026-11-03 08:08:42 +02:00
Aman Gupta ee17f293b6 kv-cache : never return 0 kv_rows for sequences without cells (#31570)
A cache that shares the cells of another one (f.ex. the gemma4 assistant) can build a graph before its sequences have any cells, f.ex. at the server warmup. get_n_kv_rows() then returned 0, which means that kv_rows are not used, so the graph fell back to the dense KQ mask and the scheduler re-planned for it. Later graphs with larger masks then aborted with GGML_SCHED_DEBUG_REALLOC=1.
2026-11-03 07:51:53 +02:00
Aman Gupta 2cbca389f5 dflash: copy hidden features via device copies (#31506) 2026-11-03 09:02:32 +08:00
Maphist0andSigbjørn Skjæret e4b6c8ca6a model: add Clover support (#31306)
* model : add AR-only Clover MoE support

Add GGUF conversion, model loading, and causal hybrid attention for Clover, reusing shared attention, recurrent-state, MoE, and LoRA helpers.

* Update tests/test-llama-archs.cpp

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

* convert : filter Clover tensors using filter_tensors

* clover : use the standard ssm_a tensor name

---------

Co-authored-by: Sigbjørn Skjæret <sigbjorn.skjaeret@huggingface.co>
2026-11-02 18:37:27 +01:00
Aman GuptaandGeorgi Gerganov 1b6935b227 llama : flash_attn_ext_rows - more efficient unified kv for multi-seq (#29510)
* llama : use llm_attn_mask and kv_rows for unified attention

Assisted-by: dsh:llama.cpp/ggml-org/DeepSeek-V4-Flash-Vision-Exp-GGUF (Mac Studio, M2 Ultra, 192GB)

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

* cont : optimize `get_seq_cells`

* llama : build the kv_rows lists from per-sequence cell bitmaps

Keep a bitmap over the cells for each sequence (the transpose of `seq`), updated in seq_pos_inc/dec, and build the cell lists by walking its set bits. The lists stay in cell order, cells shared by several sequences go to every list, and nothing has to be rebuilt when cells are removed.

Also leave out of the lists the cells that the SWA hides from every token of the slice (standard and chunked SWA).

* cont : make get_kv_rows_p_min private

* cont : minor metal

* llama : use kv_rows for every ubatch of a unified cache

A ubatch made of equal-length runs with a single sequence each (f.ex. the drafts of several slots) now splits into per-sequence slices.

Any other ubatch (interleaved, uneven or coupled sequences) becomes a single slice over the cells of all its sequences instead of falling back to the dense KQ mask, so the graph keeps the topology of the reserved graph.

* llama : merge get_kv_rows_seq_ids and get_kv_rows_p_min into get_kv_rows_slices

Assisted-by: dsh:llama.cpp/ggml-org/GLM-5.3-Flash-GGUF (Mac Studio, M5 Ultra, 256GB)

---------

Co-authored-by: Georgi Gerganov <ggerganov@gmail.com>
2026-11-02 18:13:33 +02:00
Niklas Wenzel cd8427d018 context : simplify setters (#29730) 2026-11-02 14:13:31 +02:00
Niklas Wenzel 79bce0c881 tests : enable Gemma 4 in test-llama-archs (#31401)
* tests : enable Gemma 4 in `test-llama-archs`

* tests : use GQA for Gemma 4 in `test-llama-archs`
2026-10-31 23:55:02 +01:00
Tarek Dakhran bc428e407a llama : enable tensor split for LFM3 (#31431)
Assisted-by: Claude Opus 5.5
2026-10-31 23:40:50 +01:00
Niklas Wenzel 0138937e3f llama : fix rotation of attention tensors in layers with shared KV (#31406)
* llama : fix rotation of layers that reuse another layer's KV cache

* llama : use KV source rotation only for attn_* tensors
2026-10-31 23:30:49 +01:00
Anthony Maltsev ddf48f85c9 reshaping so mm on 2d instead of 3d tensor for lfm sc (#30831) 2026-10-31 20:34:57 +01:00
Tarek Dakhran 889f2ec96d quant : allow quantization for LFM3-Audio codec decoder (#31386)
* quantize : quantize the LFM3-Audio codec decoder, it is Q4_0 QAT

* quantize : drop the QAT note from the LFM3-Audio comment
2026-10-30 11:52:59 +01:00
Lorenzo Bona 344846b4fc gemma3n, gemma4 : create token_embd before the tied output (#31394)
The duplicated output looks up token_embd in its buffer context, so
token_embd must exist first. With the old order a load where input and
output share a buffer type (CPU-only, -ngl 0) held the table twice:
gemma-4-E2B-it Q4_K_M, no mmap, CPU model buffer 3163.73 -> 2947.73 MiB.
2026-10-30 09:36:27 +01:00
Anton IandPascal fb5997274e kv-cache : keep rotation constants on the GPU (#31073)
* kv-cache : keep rotation constants on the GPU

Assisted-by: Codex

* tests : verify resident KV rotation buffers

Assisted-by: Codex

* kv-cache : document remaining multi-device rotation work

Assisted-by: Codex

* kv-cache : support per-layer rotation buffers

Assisted-by: Codex

* kv-cache : fix hybrid SWA rotation inputs and mixed placement check

Assisted-by: Codex

* models : use per-layer rotation in mistral-large4 and dflash

* tests: drop the rotation buffer tests

---------

Co-authored-by: Pascal <admin@serveurperso.com>
2026-10-30 10:15:06 +02:00
Pascal ab3768cdc2 memory: refuse rollbacks past the snapshots of the last ubatch (#31368)
* memory: refuse rollbacks past the snapshots of the last ubatch

The recurrent memory and the DSV4 cache accepted any rollback up to
n_rs_seq, whether or not the last ubatch wrote that snapshot. Each now
records the deepest rollback its last ubatch guarantees and seq_rm
refuses anything past it:

- recurrent: n - 1 for a ubatch of n <= n_rs_seq + 1 tokens, 0 for a
  longer one, which only guarantees the final state
- DSV4: min(n, n_rs_seq), the snapshots are rebuilt from the state
  before the ubatch

The dead rollback block of the rollback test becomes the assertion it
asked for, and rollback_bound checks the exact bound of each memory
after ubatches of 1, 2, n_rs_seq + 1 and n_rs_seq + 2 tokens.

* tests: address review from ggerganov

Detect architectures with llama_arch_from_string in comp_padding and
arch_reserves_single_seq, as in rollback_bound.
2026-10-30 10:11:58 +02:00
Georgi Gerganov 418f905a4e ggml : add batched tensor copy API (#30881)
* ggml : add generic batched tensor copy API

Assisted-by: pi:llama.cpp/ggml-org/DeepSeek-V4-Flash-Vision-Exp-GGUF (Mac Studio, M2 Ultra, 192GB)

* llama : use batched tensor copy API in context io

Assisted-by: pi:llama.cpp/ggml-org/DeepSeek-V4-Flash-Vision-Exp-GGUF (Mac Studio, M2 Ultra, 192GB)

* ggml : add batched tensor copy API

Assisted-by: pi:llama.cpp/ggml-org/DeepSeek-V4-Flash-Vision-Exp-GGUF (Mac Studio, M2 Ultra, 192GB)

* llama : batch device io tensor copies

Assisted-by: pi:llama.cpp/ggml-org/DeepSeek-V4-Flash-Vision-Exp-GGUF (Mac Studio, M2 Ultra, 192GB)

* ggml : add CUDA support for batched tensor APIs

Assisted-by: pi:llama.cpp/ggml-org/DeepSeek-V4-Flash-Vision-Exp-GGUF (Mac Studio, M2 Ultra, 192GB)

* ggml : remove async tensor_n APIs and add null buffer hooks

Assisted-by: pi:llama.cpp/ggml-org/DeepSeek-V4-Flash-Vision-Exp-GGUF (Mac Studio, M2 Ultra, 192GB)

* ggml : fix batched tensor 2D detection

Assisted-by: pi:llama.cpp/ggml-org/DeepSeek-V4-Flash-Vision-Exp-GGUF (Mac Studio, M2 Ultra, 192GB)

* ggml : group batched tensor operations by buffer type

Assisted-by: pi:llama.cpp/ggml-org/DeepSeek-V4-Flash-Vision-Exp-GGUF (Mac Studio, M2 Ultra, 192GB)

* cont : minor indentantion

* cont : const correctness

* tests : add mock buffer type tests for batched tensor API

Restructure test-backend-tensor to run a unified test set on each backend buffer type and on an ad-hoc mock buffer type. The mock set exercises the dispatch logic in ggml_backend_tensor_{set,get,copy}_n: hook invocation, per-tensor fallback, cpy_tensor_n all-or-nothing, and buft grouping.

Assisted-by: dsh:llama.cpp/ggml-org/DeepSeek-V4-Flash-Vision-Exp-GGUF (Mac Studio, M2 Ultra, 192GB)

* cont : add TODO
2026-10-29 20:52:17 +02:00
Oliver SimonsandSigbjørn Skjæret 68d20b884c Quant: OCP FP8 E4M3 support (#27512)
* WIP OCP FP8 E4M3 support

* Fix python type-checker

* Materialize input_scales also for FP8 when converting

* Fix conversion of https://huggingface.co/RedHatAI/Qwen3.8-27B-NVFP4

1. Per row FP8 wrongly fell to NVFP4 conversion, as NVFP4 checked on
   dims alone. Also check on dtype for FP8
2. Need to reshape FP8 QKV projections scales in the same way that
   weights are reshaped

* Allow remaining FP8 tensors to be dequantized during conversion

* Allow tensor-strategy for FP8-only compressed-tensor ckpts

* Fix and test gguf-py behavior for FP8_E4M3

* Disallow QKV-fusion for FP8 temporarily

* Fix conversion of fused qkvz & FP8 for Qwen3-Next

* Avoid fusing FP8 for --fuse-gate-up-exps as we may have different ws

* Emit warnings when not fusing FP8 during conversion

* More FP8 conversion fixes

* Add supported for strided copies to/from FP8 in CPU

* remove unneeded scratch buffer from FP8 cpu cpy/dup path

* add explanatory comments and TODOs

* Add dedicated encoder and decoder tests for FP8

* Whitelist ggml_op_get_rows in metal instead of blacklisting

required to fix failing CI, and white-listing should be more
future-proof than adding FP8 to the black-list

https://github.com/ggml-org/llama.cpp/actions/runs/38433916498/job/115422880633?pr=27512

* Fix more conversion bugs

* Defer weight materialization during conversion

* Blacklist fp8 in metal backend instead of white-list

* Adjust indents

* Fix lint + execute tests

* Add fast CPU kernel for F8_E4M3

* Conversion fixes

* Fix remaining gaps

* Enable tied weights with shared scales

* Make mandatory-F32 rules apply to FP8 during conversion

* Fix python type checks

https://github.com/ggml-org/llama.cpp/actions/runs/39765670052/job/119780379595?pr=27512

* Fix kaleidai unit-test failures

https://github.com/ggml-org/llama.cpp/actions/runs/39765670125/job/119780383150

* Relax numeric precision constraints for pre-check

* Leave fast-FP8 CPU on RISC-V for follow-up

We already exclude arm from the fast-path

* Allow output scale tensors to be located in any gguf shard

* Move test_conversion_fp8.py test execution to behavior-check.yaml

* Apply batched suggestions from code review

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

* Move mixed compressed-tensor logic into dequant_fp8 + add unit-tests for it

* Install all requirements also on Windows for behavior-check

Fixes https://github.com/ggml-org/llama.cpp/actions/runs/40535852590/job/122300758223?pr=27512

* Unmap lazy tensors before clean-up in Windows

Should fix https://github.com/ggml-org/llama.cpp/actions/runs/40539797824/job/122314782459?pr=27512

* LF line endings

https://github.com/ggml-org/llama.cpp/actions/runs/40539798168/job/122314791118?pr=27512

* Extract FP8 tests from llama-arch to its own test-fixture

See https://github.com/ggml-org/llama.cpp/pull/27512#discussion_r4400628221

* Update gguf-py/gguf/scripts/gguf_convert_endian.py

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

---------

Co-authored-by: Sigbjørn Skjæret <sigbjorn.skjaeret@huggingface.co>
2026-10-29 18:14:37 +01:00
Xuan-Son Nguyen 78feec062d model: mistral le chonk (#31376)
* llama: extend DSA and sdpark to new model

* fix ci
2026-10-29 15:43:33 +01:00
Aman Gupta 0ac1a7fd35 llama : reuse the tensor staging buffer and skip the direct-io bounce buffer (#31321) 2026-10-29 19:22:43 +08:00
Giuseppe Scrivano abb76ab380 model : Support Limite 1B - Violetto (#29433)
* gguf-py : add Limite arch

Assisted-by: Claude Opus 5.5
Signed-off-by: Giuseppe Scrivano <giuseppe@scrivano.org>

* convert : add Limite support

Assisted-by: Claude Opus 5.5
Signed-off-by: Giuseppe Scrivano <giuseppe@scrivano.org>

* llama : add Limite support

Assisted-by: Claude Opus 5.5
Signed-off-by: Giuseppe Scrivano <giuseppe@scrivano.org>

---------

Signed-off-by: Giuseppe Scrivano <giuseppe@scrivano.org>
2026-10-29 10:13:41 +02:00
Yonghye Kwon 838093f192 llama : remove simple batch splitting for DSV4 (#31300)
* llama : fix DSV4 compressor padding for simple batches

Count actual tokens per sequence when padding compressor writes. Add a graph reuse regression test for odd-sized DeepSeek V4 and V4.1 batches.

Assisted-by: Codex

* tests : cover multiple sequences in DSV4 padding regression

Exercise one and two sequences at different positions while keeping a fixed ubatch shape.

Assisted-by: Codex

* llama : remove simple batch splitting for DSV4

Use the existing sequence splitting to keep n_seq_tokens consistent with
the compressor padding calculation. Extend the graph reuse regression to
6-token batches.

Assisted-by: Codex
2026-10-28 23:25:47 +08:00
Georgi Gerganov f1a7dea99a tests : adjust constants to reduce irrelevant warnings (#31254) 2026-10-28 11:26:38 +02:00
Tarek DakhranandXuan Son Nguyen d05706b4f3 model : add LiquidAI/d1-2B and LiquidAI/d1-omni-2B (#31275)
* model : add LiquidAI/d1-2B and LiquidAI/d1-omni-2B

Assisted-by: Claude Opus 5.5

* convert : keep the speech task templates out of the d1 GGUFs

Assisted-by: Claude Opus 5.5

* small refactoring: no_batching

---------

Co-authored-by: Xuan Son Nguyen <son@huggingface.co>
2026-10-27 23:36:32 +01:00
Tarek DakhranandXuan Son Nguyen 19ff817fb2 model : add LiquidAI/LFM3-Audio-2B and LiquidAI/LFM3-Omni-2B (#31257)
* model : add LiquidAI/LFM3-Audio-2B and LiquidAI/LFM3-Omni-2B

Co-authored-by: Xuan Son Nguyen <son@huggingface.co>
Assisted-by: Claude Opus 5.5

* convert : read the LFM3-Audio config.json as released

A transformers that knows lfm3_audio rewrites the config into another layout (no lfm/encoder keys).

Assisted-by: Claude Opus 5.5

* mtmd : disable flash attention for the LFM3-Audio depthformer

Assisted-by: Claude Opus 5.5

* tts : add a voice option for LFM3-Audio and LFM3-Omni

The model card offers a female and a male voice; --tts-voice for llama-tts, "voice" for the server /tts.

Assisted-by: Claude Opus 5.5

* clean up template handling

* convert : detect LFM3-Audio and LFM3-Omni by model_type

Assisted-by: Claude Opus 5.5

* move banned tokens --> logits bias --> mtmd_helper

* leave mtmd-cli unchanged

* no more template_asr_system

* left over

* llama : keep the LFM3-Audio encoder convs and codec unquantized

Assisted-by: Claude Opus 5.5

* common, mtmd : shorten comments

Assisted-by: Claude Opus 5.5

---------

Co-authored-by: Xuan Son Nguyen <son@huggingface.co>
2026-10-27 18:47:14 +01:00
19cce69199 model : add LiquidAI/LFM3-1.7B and LiquidAI/LFM3-VL-2B (#31252)
* model : add LiquidAI/LFM3-1.7B and LiquidAI/LFM3-VL-2B

Co-authored-by: Sigbjørn Skjæret <sigbjorn.skjaeret@huggingface.co>
Co-authored-by: Xuan-Son Nguyen <son@huggingface.co>
Assisted-by: Claude Opus 5.5

* tests : set the SWA rope base of the LFM3 test model

Assisted-by: Claude Opus 5.5

* llama : enable recurrent state rollback for LFM3

Assisted-by: Claude Opus 5.5

* Apply batched suggestions from code review

Co-authored-by: Sigbjørn Skjæret <sigbjorn.skjaeret@huggingface.co>
Co-authored-by: Tarek Dakhran <github@dakhran.com>

* gguf-py : align the lfm3vl mapping comments

Assisted-by: Claude Opus 5.5

* llama : fix CRLF line endings from the review suggestions

Assisted-by: Claude Opus 5.5

* tests : add the LFM3 Metal fusion baseline

Assisted-by: Claude Opus 5.5

* llama : keep the LFM3-VL depthwise conv kernels in F32 when quantizing

Assisted-by: Claude Opus 5.5

---------

Co-authored-by: Sigbjørn Skjæret <sigbjorn.skjaeret@huggingface.co>
Co-authored-by: Xuan-Son Nguyen <son@huggingface.co>
2026-10-27 17:06:40 +01:00
Aman GuptaandGeorgi Gerganov 27b7a8a009 llama : build the k-pool sparse selection independently of the kq_mask shape (#31243)
* llama : build the k-pool sparse selection independently of the kq_mask shape

qwen4exp (QSA) and glm5-next pick cache cells by index and no longer read the kq_mask values, but they still shaped the selection like the dense kq_mask. Build the selection as [n_kv, n_tokens] over the cache cells instead, and split it into the streams of the K view right before the attention.

This lets the dense mask change its layout (e.g. per-sequence row lists) without touching the sparse layers.

* improve comment

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

* llama : drop the unused build_fused_top_k_mask args, review nits

---------

Co-authored-by: Georgi Gerganov <ggerganov@gmail.com>
2026-10-27 21:57:45 +08:00
Alexey Kopytko d78f40d9d3 glm5-next : QSA without kq_mask (+ DSA) (#31009)
* glm5-next : QSA without kq_mask

* DSA: dsa top-k mask change tested with layers 0,3 of DeepSeek-V3.2
2026-10-27 18:54:36 +09:00
Georgi Gerganov 7a2e5aba90 llama : remove obsolete n_pad from memory instances (#31200)
Every caller passes n_pad = 1 and the only two users clamp it to 256.
Drop the parameter, the members and the trivially true kv_size assert.

Assisted-by: dsh:llama.cpp/ggml-org/Qwen3.8-Flash-Next-GGUF (MacBook, M5 Max, 128GB)
2026-10-27 07:53:44 +02:00
begcdn 961c376a2d llama : enable moe amax rescaling for laguna (#31147)
* llama : enable moe amax rescaling for laguna

Assisted-by: Devin

* llama : declare laguna sm tensor support
2026-10-26 23:04:19 +01:00
Georgi Gerganov a4699400e3 llama : move no_alloc out of llama_hparams (#31141)
* llama : move no_alloc out of llama_hparams

Assisted-by: dsh:llama.cpp/ggml-org/GLM-5.3-Flash-GGUF (2x DGX Spark (RDMA, TP=2))

* cont : clean-up

* cont : clean-up

Assisted-by: dsh:llama.cpp/ggml-org/GLM-5.3-Flash-GGUF (2x DGX Spark (RDMA, TP=2))

* tests : add a no_alloc case to test-llama-context

Assisted-by: dsh:llama.cpp/ggml-org/GLM-5.3-Flash-GGUF (2x DGX Spark (RDMA, TP=2))
2026-10-26 16:21:25 +02:00
Georgi Gerganov 4b851fbee8 models : replace ggml_tensor** out-params with output structs (#31142)
The hyper-connection and DSA graph builders wrote their extra outputs
(post / comb gates, carried top-k selections, inject gates) through
ggml_tensor** out-params. Return them in a small struct next to the
primary output instead.

Assisted-by: dsh:llama.cpp/ggml-org/Qwen3.8-Flash-Next-GGUF (MacBook, M5 Max, 128GB)
2026-10-26 16:19:47 +02:00
Georgi Gerganov fd8f8451a2 tests : refactor test-llama-archs to not link internal libllama symbols (#30880)
test-llama-archs included src/llama-arch.h and src/llama-model-saver.h,
which prevented building on Windows with shared libraries since the
used symbols are not exported from the DLL.

- declare llama_arch_* wrappers around the internal llm_ arch query
  functions in the llama-ext.h staging header, together with
  llama_arch_kv_name() and llama_arch_supports_save()
- slim llama_model_saver down to the private implementation of
  llama_model_save_to_file(): the fixture-writer half used by the test
  moves into tests/test-llama-archs.cpp as a local kv_writer
- replace the roundtrip saver usage with the public
  llama_model_save_to_file() writing to a temp path (also removes the
  tmpfile() fd leak and its Windows workaround)
- move test-llama-archs, test-generate-models, test-llama-context and
  test-fusion out of the NOT WIN32 OR NOT BUILD_SHARED_LIBS guard

Assisted-by: pi:llama.cpp/ggml-org/GLM-5.3-Flash-GGUF (2x DGX Spark (RDMA, TP=2))
2026-10-26 12:16:02 +02:00
Aman GuptaandGeorgi Gerganov a074215352 deepseekv4.1: follow-up review fixes (#31137)
* kv-cache : use ggml_backend_tensor_copy to refill the DSV4 K mirrors

* llama : move the Engram pad id to llama_hparams and the token map to llama_model

The saver no longer needs the deepseek41 model class.

* deepseek41 : drop the F32 src1 request on the HC mixes matmul

* ggml : add softplus TODOs in the backends that still use log(1 + exp(x))

* convert : make the DSV4/DSV4.1 modify_tensors generators, drop the redundant vision filter

* deepseek41 : clean up comments

* deepseek41 : drop the F32 src1 requests on the router and indexer matmuls

* convert : use format_tensor_name directly in DeepseekV4Model

* s/comp_kv_src/comp_il_src/g

* cont : fix TODOs for log1p

---------

Co-authored-by: Georgi Gerganov <ggerganov@gmail.com>
2026-10-26 15:52:02 +08:00
48c9f08b7b common, grammar: add recursive depth limit, make it a requirement (#31153)
* common, grammar: add recursive depth limit, make it a requirement

* catch more recursive functions

* common : tidy up peg parser changes

* grammar : reject candidates in a loop, without a depth limit

The depth limit in llama_grammar_reject_candidates_for_stack counted
the code points of a vocab token, so any token of 256 chars or more
was always rejected under a grammar, such as the 512-space token of
GLM-4.7-Flash or the long space tokens of Laguna-S-2.1.

The rejection now walks the code points with its own stack of levels
instead of recursing: there is no limit and no stack risk, whatever
the length of a token from the GGUF, and the masks match master.

* grammar : track rejected candidates as indices into the input

At a given code point a candidate only differs from its input by the
position of its code point pointer, so the levels now carry indices
into the input candidates instead of copies. The first level needs no
special case and the rejects map back to the input directly.

Grammar sampling gets faster than master on GLM-4.7, Laguna, Qwen3.6
and gemma-4 with the json, list and spaces grammars, same masks.

* nit : shorten the comment of llama_grammar_reject_step

* common : add depth_guard for schema walks

* grammar : move GRAMMAR_RECURSIVE_LIMIT into the .cpp

* common : reduce recursion limit for peg parser

* improve

---------

Co-authored-by: Alde Rojas <hello@alde.dev>
Co-authored-by: Pascal <admin@serveurperso.com>
2026-10-25 23:53:08 +01:00
Aman Gupta 2280f0fd16 qwen: merge tensors on convert (#31028)
* qwen: merge tensors on convert

* convert: use generator functions

* qwen4exp: remove conts and pre-scale
2026-10-25 15:23:22 +02:00
Trunk Trick 63412d8737 llama: fix exhausting CPU and MEM caused by dry sequence breaker (#31035) 2026-10-25 12:31:09 +01:00
Anthony Wang 5c965fd8ca llama-quant : fix 4D tensor quantization using ggml_nrows() and name-based expert detection (#30820)
* llama-quant : fix quantizing 4D tensors

nrows_total was ne[1]*ne[2], so for a 4D tensor only the first ne[2]
slices were quantized and the rest of the rows were dropped. Use
ggml_nrows(tensor) to count all rows, including ne[3].

Fixes #29916

* llama-quant : detect expert tensors by name and fix imatrix size for 4D tensors

The MXFP4_MOE path used ne[2] > 1 to detect MoE expert tensors. This is
wrong for non-expert tensors with 3 or more dims, e.g. MLA projections
(BailingMoE3) and 4D conv kernels. Use the tensor name instead: all
tensors used by GGML_OP_MUL_MAT_ID are named *_exps or *_chexps. This
also removes the BailingMoE3 special case.

The imatrix holds one vector of ne[0] values per 2D slice, so the
expected size is ne[0]*ne[2]*ne[3], not ne[0]*ne[2]. This matches how
the imatrix is indexed per row for 4D tensors in llama_tensor_quantize_impl.

Assisted-by: Claude Opus 5.5
2026-10-25 13:28:38 +02:00
Hrushikesh Ramilla f1168ac417 grammar: limit recursion depth to prevent stack overflow (#29495) (#29755) 2026-10-25 12:10:10 +01:00
Georgi Gerganov fa3645feb5 llama : respect no_alloc in all memory caches (#31087)
* llama : respect no_alloc in memory caches

Assisted-by: dsh:llama.cpp/ggml-org/DeepSeek-V4-Flash-Vision-Exp-GGUF (Mac Studio, M2 Ultra, 192GB)

* llama : TODO: move no_alloc out of llama_hparams

Assisted-by: dsh:llama.cpp/ggml-org/DeepSeek-V4-Flash-Vision-Exp-GGUF (Mac Studio, M2 Ultra, 192GB)

* llama : TODO: move no_alloc out of llama_hparams

Assisted-by: dsh:llama.cpp/ggml-org/DeepSeek-V4-Flash-Vision-Exp-GGUF (Mac Studio, M2 Ultra, 192GB)
2026-10-25 11:56:04 +02:00
Aman GuptaandDaniel Han 74bca39500 model: DSV4.1 flash (#31086)
* model : add DeepSeek-V4.1-Flash (deepseek41)

* deepseek41 : remove MXFP8, convert the Engram table to BF16

* deepseek41 : compute the compressed-KV read layers once in the cache

* deepseek41 : use get_cells() for the shared-history check, drop llama_kv_cache::seq_pos_tok_le

* deepseek41 : gather the Engram rows in-graph like qwen4exp, drop the host gather

* deepseek41 : keep the Engram hash constants in hparams.ple_*, one module per Engram layer

* deepseek41 : drop the LLAMA_DSV41_KV_MIRROR toggle

* deepseek41 : keep il last in build_hc_mixes

* deepseek41 : sort the top-k rows before the sparse gather

* tests : add deepseek41 to the Metal fusion baseline

---------

Co-authored-by: Daniel Han <danielhanchen@gmail.com>
2026-10-25 09:43:51 +08:00
Satyam Thakurandorigin-labs-ai 73adf89e38 minimax-m3 : forward quant scales to the graph (fixes #31063) (#31097)
The generic scale pass in llama-model.cpp loads the per-expert
NVFP4 scale2 tensors (ffn_{up,gate,down}_exps_s) and the per-tensor
MXFP8 scales for the dense FFN and the shared expert, but
minimax-m3.cpp never forwarded any of them to the graph.

build_moe_ffn() defaults these to nullptr, so build_lora_mm_id()
skipped its w_s branch entirely and the routed-expert weights came
out global_scale times too large, swamping the residual (reported
G ~= 15123 for M3.1 Flash) and producing token soup.

Forward all scale tensors at the three call sites, matching the
pattern used by flint and the other NVFP4-capable MoE models.

Co-authored-by: origin-labs-ai <origin-labs-ai@users.noreply.github.com>
2026-10-24 16:09:02 +02:00
Georgi Gerganov 3032209758 llama : deny -sm tensor for arches with moe amax rescaling (#31081)
The amax POOL_1D reduction spans the whole matmul row, so a device cannot
compute it from an axis-0 split. The Meta split-state case reported
MIRRORED, which gives each device a partial max of its own slice. Drop
that case and deny -sm tensor for the arches which build the amax graph,
see [TAG_MOE_AMAX_SCALE].

Assisted-by: dsh:llama.cpp/ggml-org/Qwen3.8-Flash-Next-GGUF (MacBook, M5 Max, 128GB)
2026-10-24 13:48:32 +03:00
Hans 9ae73e68f5 mtmd : add Soprano TTS support (#28607)
Add Soprano-80M and Soprano-1.1-80M (https://huggingface.co/ekwek/Soprano-1.1-80M)
as an audio generation model: a Qwen3 backbone plus a Vocos decoder that runs
as a gen-only mmproj.

- convert: detect Soprano checkpoints and export the backbone as qwen3 and the
  decoder.pth Vocos weights as the mmproj
- vocab: add the soprano pre-tokenizer (lowercase, whitespace collapse,
  per-digit split, [UNK] fallback)
- mtmd: add the Vocos graph and ISTFT, allow loading a gen-only mmproj for
  Soprano and include the gen-audio context in mtmd_get_memory_usage
- mtmd-helper: add the Soprano pipeline; the decoder runs over the whole
  utterance, so stream mode sends the audio in one chunk after flush()
- tts: document Soprano usage

Assisted-by: Codex
Assisted-by: Claude Opus 5.5
2026-10-24 01:09:50 +02:00
Christian Loehnert 06230cb877 allows the usage of 4-10 gpus for stepfun and laguna in tensor parallel (#24554) 2026-10-23 19:56:38 +02:00