Commit Graph
14 Commits
Author SHA1 Message Date
Xuan-Son Nguyen 6023692b1e llama: fix kv shifting, add tests, add per-model build_rope (#31683)
* llama: fix kv shifting and add tests

* llama: add graph build_rope (#31693)

* llama: add graph build_rope

* improve

* migrate new models

* update review skill

* fix ci

* add --kv-shift to CI
2026-11-05 21:10:36 +01:00
Xuan-Son Nguyen e0f96a1a12 llama: refactoring: move model to class member of llm_graph_context (#31733) 2026-11-05 15:41:03 +01:00
Pascal cb55ffe6e2 graph: skip the LM head for pooled embeddings (#30445)
* graph: skip the LM head for pooled embeddings

A context that extracts pooled embeddings outputs one vector per
sequence, but every model still expanded its LM head, so the graph
computed n_vocab logits per token and the context kept a host buffer
for them.

The models now only declare t_logits, and set_outputs() expands it
unless the context extracts embeddings with a pooling type other than
NONE. Since the toggle changes the graph shape for a pooled context,
set_embeddings() schedules a re-reserve in that case.

* graph: drop the redundant t_embd expand

Address review from @CISC: with embeddings on, build_pooling() already
expands its output for every pooling type, and a decoder graph without
embeddings does not read t_embd, so set_outputs() only expands
t_logits.

* graph: add build_logits() for the LM head output

Address review from @ggerganov: the LM head is expanded by a dedicated
build_logits(), called from llama_model::build_graph() like
build_pooling(), and set_outputs() only marks the outputs again.

* graph: call build_logits() before build_sampling()

Address review from @ggerganov: build_sampling() reads t_logits, so the
LM head is built first. A pooled context then skips the sampling layers
too, since it has no logits to sample from.

* nit
2026-10-13 16:16:14 +02:00
Pascal 542348a35c Model-Saver: Write the SWA pattern, 15 more architectures roundtrip (#29042)
* llama: read the SWA pattern as a period or a per-layer array

Add llama_model_base::load_swa_pattern(), which reads
sliding_window_pattern either as one flag per layer or as a period
expanded by set_swa_pattern(), and use it in every loader that reads
the key as a period.

These loaders silently ignored an array and applied their default
period, although the converters of olmo2, gemma3n and exaone4 write
arrays. The published GGUFs match the defaults, so their outputs do
not change. The loaders that already accepted both forms lose their
duplicated scalar-then-array block, and use their declared default
period when the key is absent.

* model-saver: write the SWA pattern and the MLA SWA geometry

Write sliding_window_pattern as one flag per layer, nextn layers
included, for every model using SWA. The array is never collapsed to
a scalar, since the loaders read a scalar as a period.

Also write the MLA key/value lengths and KV LoRA rank of the SWA
layers, required by dots3note.

This enables the saver for plamo3, gemma3, cohere2, cohere2moe,
olmo2, exaone-moe, afmoe, mimo2, spark2_5, muse-glimmer, mellum,
laguna, granite_swa, dots3note and maple, all passing the bit-exact
roundtrip of test-llama-archs.
2026-09-18 15:20:03 +02:00
Sigbjørn Skjæret 9d817213a0 model : load hparams.n_layer_nextn before n_layer() calls (#28159)
* load hparams.n_layer_nextn before n_layer() calls

* remove duplicate loads
2026-09-01 13:55:45 +02:00
Junmo Kim 14e78ddef7 model : fix SWA not being enabled for EXAONE 4.5 (#26848)
* model : fix SWA not being enabled for EXAONE 4.5

load_arch_hparams tests `hparams.n_layer() == 64` before
LLM_KV_NEXTN_PREDICT_LAYERS has been read. n_layer() returns
n_layer_all - n_layer_nextn and n_layer_nextn defaults to 0, so a GGUF
carrying the MTP head (block_count=65, nextn=1) evaluates to 65 and the
whole SWA block is skipped. The model type switch further down in the
same function reads 64, because by then the key has been loaded.

n_swa is still filled in by the unconditional get_key below the block, so
llama_model_n_swa() reports 4096 and the logs look correct while only
swa_type stays LLAMA_SWA_TYPE_NONE.

This affects the official LGAI-EXAONE GGUF release as well. EXAONE 4.0 has
no MTP head, so block_count is 64 there and the check matches.

* model-loader : skip TENSOR_SKIP tensors in the metadata-only path

create_tensor asserts on a null buffer type when building from metadata
alone, but buft_for_tensor returns null by design for tensors marked
TENSOR_SKIP, which is how architectures with nextn/MTP layers mark theirs.
Those models cannot be constructed by llama_model_init_from_user at all.

The file-backed path below already returns nullptr for the same tensors, so
callers see the same thing either way.

* tests : cover exaone4 hparams ordering

Builds a synthetic exaone4 model with the layout the shipped EXAONE 4.5
GGUFs use (block_count 65 + nextn 1). The swa_type check is the one that
catches the ordering bug; the n_layer_nextn and n_layer() checks only tell
a broken fixture apart from a real regression.

Fails before the ordering fix with "swa_type is not STANDARD", passes after.

* Revert "tests : cover exaone4 hparams ordering"

This reverts commit d2f3bafeee591ad691396b2708de4baef3aaf602.

* Revert "model-loader : skip TENSOR_SKIP tensors in the metadata-only path"

This reverts commit aecb9bc0c7896b52afbc43921a1f572aa7b5e53c.
2026-08-11 07:20:17 +03:00
Georgi GerganovandSigbjørn Skjæret 7acb4e8cd2 hparams : refactor hparams.n_layer (#24060)
* hparams : refactor hparams.n_layer

* cont : remove `n_layer_kv()`, use n_layer_all instead

* cont : type consistency

* pi : update SYSTEM.md

* models : fix Step3.5 MTP

* cont : remove duplicate switch cases

* cont : explicitly set `false` to extra layers for `is_swa` and `is_recr`

* cont : fix nextn layer count handling

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

---------

Co-authored-by: Sigbjørn Skjæret <sigbjorn.skjaeret@scala.com>
2026-06-05 11:09:36 +03:00
Junwon HwangandLG-AI-EXAONE 48b88c3b00 model: Add EXAONE 4.5 implementations (#21733)
* Add EXAONE 4.5 and Add GQA for MMproj

* mtmd: EXAONE 4.5 vision markers and projector path

EXAONE 4.5 uses <vision> and </vision> for image boundaries; Qwen keeps
<|vision_start|> and <|vision_end|>.

Route EXAONE 4.5 through the Qwen2.5-VL-style encode path (window attention
pattern, optional mmproj input norm). Update exaone4_5 projector weights and
convert_hf_to_gguf for mmproj export.

* mtmd: load EXAONE4 nextn tensors correctly

Align EXAONE4 tensor registration with EXAONE_MOE for NextN/MTP slots and avoid skip-flag propagation on duplicated rope_freqs so model loading succeeds for EXAONE 4.5 GGUF.

* Minor fixes

* Address PR feedback

* Address PR feedback

* Fix EXAONE after merge

* Fix EXAONE 4.5 conversion

* Address PR feedback

* Refactor EXAONE 4.5 conversion

* Address PR feedback

* Fix unintended deletion

* Minor fix

---------

Co-authored-by: LG-AI-EXAONE <exaonemodels@lgresearch.ai>
2026-06-01 11:48:53 +02:00
ynankani 42928bc14d model : NvFP4 quantized LM head support (#23046)
* NvFP4 quantized LM head support

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

* Address review commnets

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

* Add assert for NvFp4 lm head and tied embeddings

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

* Address review commnets

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

* Create output_s tensor only when LM head NvFp4

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

---------

Signed-off-by: ynankani <ynankani@nvidia.com>
2026-05-16 11:09:27 +02:00
Xuan-Son Nguyen 994118a183 model: move load_hparams and load_tensors to per-model definition (#22004)
* git-friendly migration

* add build_graph

* nits

* exclude old code from build

* wip

* add llm_arch_model_i

* prepare downstream functions

* nits

* nits

* wip

* wip

* add back create_tensor_qkv

* fix files missing include

* enforce one llm_build per arch

* cmake: use glob

* missing model params

* nits

* wip

* wip (2)

* wip (3)

* test-llama-archs is happy

* improve switch case

* move more stuff into llm_arch_model_i

* fix downstream code

* nits

* nits (2)

* fix order

* llama_model_base

* LLAMA_LOAD_LOCALS

* small fix

* fix build errors

* auto

* rm migration script and ifdef
2026-05-04 12:36:59 +02:00
PikaPikachu 9db77a020c model : refactor QKV into common build_qkv and create_tensor_qkv helpers (#21245)
* model : refactor QKV into common build_qkv and create_tensor_qkv helpers

* model : extend build_qkv to bert/mpt/dbrx/olmo/lfm2/nemotron-h/granite-hybrid/gemma3n-iswa/t5-dec and fix wqkv_s
2026-04-16 17:41:34 +02:00
Sigbjørn Skjæret f772f6e434 model : support NVFP4 tensors for Gemma4 (#21971)
* support nvfp4 tensors for Gemma4

* add wo_s to build_attn

* add wo_s to build_attn

* fix glm4
2026-04-16 16:51:47 +02:00
Xuan-Son Nguyen 59db9a357d llama: dynamic head_dim and n_rot for SWA (#20301)
* llama: dynamic head_dim and n_rot for SWA

* also add gguf_writer wrappers

* fix build

* build_rope_shift arg reorder
2026-03-09 22:22:39 +01:00
Piotr Wilkin (ilintar)andSigbjørn Skjæret bea04522ff refactor : llama-model.cpp (#16252)
* Sqashed: llama-model.cpp refactoring

* Fix formatting of attn / ffn / ffn_moe calls

* Fix import regression / unify spacing in models.h

* totally DID NOT miss those!

* Add missing qwen3vl(moe) models

* Add missing new .cpp files to build

* Remove extra semicolons

* Editor checker

* Update src/models/models.h

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

---------

Co-authored-by: Sigbjørn Skjæret <sigbjorn.skjaeret@scala.com>
2025-10-31 23:40:23 +01:00