Georgi Gerganov 161755f29e test-llama-archs : make tensor data stdev configurable and improve help (#29133) ...

* test-llama-archs : make tensor data stdev configurable

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

* test-llama-archs : expand usage and add examples

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

* test-llama-archs : fail on unknown args and log usage

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

* test-llama-archs : add test run summary

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

* test-llama-archs : initialize Mamba ssm_a negative

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

* test-recurrent-state-rollback : report NMSE for logits mismatches

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

* test-recurrent-state-rollback : use NMSE for rollback logits checks

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

* tests : disable invalid test

* cont : adjust nmse_eps

* tests : zero DSA indexer score projection in synthetic fixtures

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

* cont : add support for `--arch` regex

* cont : alternative top-k stability

* cont : indentation

* cont : fix top-k value

* cont : consistent logs

2026-09-21 13:57:36 +03:00

..

afmoe.cpp

Model-Saver: Write the SWA pattern, 15 more architectures roundtrip (#29042)

2026-09-18 15:20:03 +02:00

apertus.cpp

hparams : refactor hparams.n_layer (#24060)

2026-06-05 11:09:36 +03:00

arcee.cpp

hparams : refactor hparams.n_layer (#24060)

2026-06-05 11:09:36 +03:00

arctic.cpp

hparams : refactor hparams.n_layer (#24060)

2026-06-05 11:09:36 +03:00

arwkv7.cpp

hparams : refactor hparams.n_layer (#24060)

2026-06-05 11:09:36 +03:00

baichuan.cpp

hparams : refactor hparams.n_layer (#24060)

2026-06-05 11:09:36 +03:00

bailingmoe2.cpp

model: add NVIDIA Nemotron-3-Puzzle-75B-A9B (NemotronHPuzzle) support (#25444)

2026-09-03 08:53:08 +02:00

bailingmoe3.cpp

models : fix GDN normalization from max to rsqrt (#28068)

2026-09-06 18:46:21 +02:00

bailingmoe.cpp

model: add NVIDIA Nemotron-3-Puzzle-75B-A9B (NemotronHPuzzle) support (#25444)

2026-09-03 08:53:08 +02:00

bert.cpp

models: clean up some dead switch branches in old models (#28669)

2026-09-10 16:09:35 +03:00

bitnet.cpp

hparams : refactor hparams.n_layer (#24060)

2026-06-05 11:09:36 +03:00

bloom.cpp

hparams : refactor hparams.n_layer (#24060)

2026-06-05 11:09:36 +03:00

chameleon.cpp

hparams : refactor hparams.n_layer (#24060)

2026-06-05 11:09:36 +03:00

chatglm.cpp

hparams : refactor hparams.n_layer (#24060)

2026-06-05 11:09:36 +03:00

clip.cpp

llama: Restore quantization of mmprojs (#26818)

2026-08-10 11:58:32 +02:00

codeshell.cpp

hparams : refactor hparams.n_layer (#24060)

2026-06-05 11:09:36 +03:00

cogvlm.cpp

hparams : refactor hparams.n_layer (#24060)

2026-06-05 11:09:36 +03:00

cohere2.cpp

Model-Saver: Write the SWA pattern, 15 more architectures roundtrip (#29042)

2026-09-18 15:20:03 +02:00

cohere2moe.cpp

Model-Saver: Write the SWA pattern, 15 more architectures roundtrip (#29042)

2026-09-18 15:20:03 +02:00

command-r.cpp

hparams : refactor hparams.n_layer (#24060)

2026-06-05 11:09:36 +03:00

dbrx.cpp

hparams : refactor hparams.n_layer (#24060)

2026-06-05 11:09:36 +03:00

deci.cpp

hparams : refactor hparams.n_layer (#24060)

2026-06-05 11:09:36 +03:00

deepseek2.cpp

convert : add --fuse-qkv flag to fuse Q/K/V into QKV during HF-to-GGUF conversion (#22780)

2026-09-07 00:47:05 +08:00

deepseek2ocr.cpp

convert : add --fuse-qkv flag to fuse Q/K/V into QKV during HF-to-GGUF conversion (#22780)

2026-09-07 00:47:05 +08:00

deepseek4.cpp

model, mtmd: fix gemma4 vision handling (#28335)

2026-09-04 12:23:27 +02:00

deepseek32.cpp

model: add NVIDIA Nemotron-3-Puzzle-75B-A9B (NemotronHPuzzle) support (#25444)

2026-09-03 08:53:08 +02:00

deepseek.cpp

model: add NVIDIA Nemotron-3-Puzzle-75B-A9B (NemotronHPuzzle) support (#25444)

2026-09-03 08:53:08 +02:00

delta-net-base.cpp

llama: refactor fused ops (#24646)

2026-07-08 18:18:09 +08:00

dflash.cpp

models : move build_arch_graph() after graph() template specialization (#28934)

2026-09-15 12:33:26 +03:00

dots1.cpp

model: add NVIDIA Nemotron-3-Puzzle-75B-A9B (NemotronHPuzzle) support (#25444)

2026-09-03 08:53:08 +02:00

dots3note.cpp

model: add NVIDIA Nemotron-3-Puzzle-75B-A9B (NemotronHPuzzle) support (#25444)

2026-09-03 08:53:08 +02:00

dream.cpp

hparams : refactor hparams.n_layer (#24060)

2026-06-05 11:09:36 +03:00

eagle3.cpp

models : move build_arch_graph() after graph() template specialization (#28934)

2026-09-15 12:33:26 +03:00

ernie4-5-moe.cpp

model : NvFP4 quantized LM head support (#23046)

2026-05-16 11:09:27 +02:00

ernie4-5.cpp

model: add NVIDIA Nemotron-3-Puzzle-75B-A9B (NemotronHPuzzle) support (#25444)

2026-09-03 08:53:08 +02:00

eurobert.cpp

hparams : refactor hparams.n_layer (#24060)

2026-06-05 11:09:36 +03:00

exaone4.cpp

Model-Saver: Write the SWA pattern, 15 more architectures roundtrip (#29042)

2026-09-18 15:20:03 +02:00

exaone-moe.cpp

Model-Saver: Write the SWA pattern, 15 more architectures roundtrip (#29042)

2026-09-18 15:20:03 +02:00

exaone.cpp

hparams : refactor hparams.n_layer (#24060)

2026-06-05 11:09:36 +03:00

falcon-h1.cpp

hparams : refactor hparams.n_layer (#24060)

2026-06-05 11:09:36 +03:00

falcon.cpp

hparams : refactor hparams.n_layer (#24060)

2026-06-05 11:09:36 +03:00

gemma2.cpp

Model-Saver: Write the SWA pattern, 15 more architectures roundtrip (#29042)

2026-09-18 15:20:03 +02:00

gemma3.cpp

Model-Saver: Write the SWA pattern, 15 more architectures roundtrip (#29042)

2026-09-18 15:20:03 +02:00

gemma3n.cpp

Model-Saver: Write the SWA pattern, 15 more architectures roundtrip (#29042)

2026-09-18 15:20:03 +02:00

gemma4-assistant.cpp

models : fix incorrect uses of get_key_or_arr (#28868)

2026-09-14 14:05:37 +03:00

gemma4.cpp

models : fix incorrect uses of get_key_or_arr (#28868)

2026-09-14 14:05:37 +03:00

gemma-embedding.cpp

Model-Saver: Write the SWA pattern, 15 more architectures roundtrip (#29042)

2026-09-18 15:20:03 +02:00

gemma.cpp

hparams : refactor hparams.n_layer (#24060)

2026-06-05 11:09:36 +03:00

glm4-moe.cpp

model: add NVIDIA Nemotron-3-Puzzle-75B-A9B (NemotronHPuzzle) support (#25444)

2026-09-03 08:53:08 +02:00

glm4.cpp

model : load hparams.n_layer_nextn before n_layer() calls (#28159)

2026-09-01 13:55:45 +02:00

glm-dsa.cpp

model: add NVIDIA Nemotron-3-Puzzle-75B-A9B (NemotronHPuzzle) support (#25444)

2026-09-03 08:53:08 +02:00

gpt2.cpp

hparams : refactor hparams.n_layer (#24060)

2026-06-05 11:09:36 +03:00

gptneox.cpp

hparams : refactor hparams.n_layer (#24060)

2026-06-05 11:09:36 +03:00

granite-hybrid.cpp

model: fix all granite family parameter counts (#28643)

2026-09-10 09:49:56 +03:00

granite-moe.cpp

model: fix all granite family parameter counts (#28643)

2026-09-10 09:49:56 +03:00

granite-swa.cpp

model: add NVIDIA Nemotron-3-Puzzle-75B-A9B (NemotronHPuzzle) support (#25444)

2026-09-03 08:53:08 +02:00

granite-switch.cpp

model : GraniteSWAForCausalLM / GraniteMoeSWAForCausalLM (#25505)

2026-08-19 16:53:31 +02:00

granite.cpp

model: fix all granite family parameter counts (#28643)

2026-09-10 09:49:56 +03:00

grok.cpp

model: add NVIDIA Nemotron-3-Puzzle-75B-A9B (NemotronHPuzzle) support (#25444)

2026-09-03 08:53:08 +02:00

grovemoe.cpp

model: add NVIDIA Nemotron-3-Puzzle-75B-A9B (NemotronHPuzzle) support (#25444)

2026-09-03 08:53:08 +02:00

hrm-text.cpp

model : add support for HrmTextForCausalLM (DFM Mimir 1B) (#27625)

2026-09-16 15:18:45 +02:00

hunyuan-dense.cpp

model: move load_hparams and load_tensors to per-model definition (#22004)

2026-05-04 12:36:59 +02:00

hunyuan-moe.cpp

model: add NVIDIA Nemotron-3-Puzzle-75B-A9B (NemotronHPuzzle) support (#25444)

2026-09-03 08:53:08 +02:00

hunyuan-vl.cpp

model : NvFP4 quantized LM head support (#23046)

2026-05-16 11:09:27 +02:00

hy-v3.cpp

model: add NVIDIA Nemotron-3-Puzzle-75B-A9B (NemotronHPuzzle) support (#25444)

2026-09-03 08:53:08 +02:00

hy-v4.cpp

model : add Tencent Hy 4 (hy_v4) preview architecture support (#28127)

2026-09-04 14:31:36 +02:00

internlm2.cpp

hparams : refactor hparams.n_layer (#24060)

2026-06-05 11:09:36 +03:00

jais2.cpp

convert : add --fuse-qkv flag to fuse Q/K/V into QKV during HF-to-GGUF conversion (#22780)

2026-09-07 00:47:05 +08:00

jais.cpp

hparams : refactor hparams.n_layer (#24060)

2026-06-05 11:09:36 +03:00

jamba.cpp

hparams : refactor hparams.n_layer (#24060)

2026-06-05 11:09:36 +03:00

jina-bert-v2.cpp

hparams : refactor hparams.n_layer (#24060)

2026-06-05 11:09:36 +03:00

jina-bert-v3.cpp

models: clean up some dead switch branches in old models (#28669)

2026-09-10 16:09:35 +03:00

kimi-k3.cpp

model : support Kimi-K3 recurrent-state rollback (#28466)

2026-09-08 11:31:48 +08:00

kimi-linear.cpp

convert : add --fuse-qkv flag to fuse Q/K/V into QKV during HF-to-GGUF conversion (#22780)

2026-09-07 00:47:05 +08:00

laguna.cpp

Model-Saver: Write the SWA pattern, 15 more architectures roundtrip (#29042)

2026-09-18 15:20:03 +02:00

lfm2.cpp

model: add NVIDIA Nemotron-3-Puzzle-75B-A9B (NemotronHPuzzle) support (#25444)

2026-09-03 08:53:08 +02:00

lfm2moe.cpp

model: add NVIDIA Nemotron-3-Puzzle-75B-A9B (NemotronHPuzzle) support (#25444)

2026-09-03 08:53:08 +02:00

llada-moe.cpp

model: add NVIDIA Nemotron-3-Puzzle-75B-A9B (NemotronHPuzzle) support (#25444)

2026-09-03 08:53:08 +02:00

llada.cpp

convert : add --fuse-qkv flag to fuse Q/K/V into QKV during HF-to-GGUF conversion (#22780)

2026-09-07 00:47:05 +08:00

llama4.cpp

Model-Saver: Write the SWA pattern, 15 more architectures roundtrip (#29042)

2026-09-18 15:20:03 +02:00

llama-embed.cpp

model: move load_hparams and load_tensors to per-model definition (#22004)

2026-05-04 12:36:59 +02:00

llama.cpp

spec: add EAGLE3 speculative decoding support (#18039)

2026-06-12 10:21:06 +03:00

maincoder.cpp

hparams : refactor hparams.n_layer (#24060)

2026-06-05 11:09:36 +03:00

mamba2.cpp

mamba2: remove hardcoded 2x expansion factor and invalid d_inner % d_state check (#23082)

2026-06-26 08:50:54 +03:00

mamba-base.cpp

fix(mamba) : make time-step projection input contiguous (#28832)

2026-09-20 07:58:14 +03:00

mamba.cpp

hparams : refactor hparams.n_layer (#24060)

2026-06-05 11:09:36 +03:00

maple.cpp

llama: add Maple 20B-A1B ternary MoE architecture (CPU) (#27000)

2026-09-14 14:04:05 +03:00

mellum.cpp

Model-Saver: Write the SWA pattern, 15 more architectures roundtrip (#29042)

2026-09-18 15:20:03 +02:00

mimo2.cpp

models : fix mimo2 swa pattern load (#28865)

2026-09-14 14:05:17 +03:00

minicpm3.cpp

model: use ggml_rope_set_offset() (#27382)

2026-08-21 18:54:29 +02:00

minicpm.cpp

hparams : refactor hparams.n_layer (#24060)

2026-06-05 11:09:36 +03:00

minimax-01.cpp

metal : single-source fusion table + fusion debug rework (#28164)

2026-09-11 12:41:54 +03:00

minimax-m2.cpp

convert : add --fuse-qkv flag to fuse Q/K/V into QKV during HF-to-GGUF conversion (#22780)

2026-09-07 00:47:05 +08:00

minimax-m3.cpp

test-llama-archs : make tensor data stdev configurable and improve help (#29133)

2026-09-21 13:57:36 +03:00

mistral3.cpp

hparams : refactor hparams.n_layer (#24060)

2026-06-05 11:09:36 +03:00

mistral4.cpp

model: move load_hparams and load_tensors to per-model definition (#22004)

2026-05-04 12:36:59 +02:00

models.h

model : add support for HrmTextForCausalLM (DFM Mimir 1B) (#27625)

2026-09-16 15:18:45 +02:00

modern-bert.cpp

Model-Saver: Write the SWA pattern, 15 more architectures roundtrip (#29042)

2026-09-18 15:20:03 +02:00

mpt.cpp

hparams : refactor hparams.n_layer (#24060)

2026-06-05 11:09:36 +03:00

muse-glimmer.cpp

Model-Saver: Write the SWA pattern, 15 more architectures roundtrip (#29042)

2026-09-18 15:20:03 +02:00

nanbeige.cpp

models : support nanbeige4.2-3B (#27730)

2026-08-27 07:55:31 +03:00

nemotron-h-moe.cpp

model : extend Nemotron MTP support (#29018)

2026-09-17 13:52:36 +02:00

nemotron-h.cpp

model : extend Nemotron MTP support (#29018)

2026-09-17 13:52:36 +02:00

nemotron.cpp

hparams : refactor hparams.n_layer (#24060)

2026-06-05 11:09:36 +03:00

neo-bert.cpp

hparams : refactor hparams.n_layer (#24060)

2026-06-05 11:09:36 +03:00

nomic-bert-moe.cpp

models: clean up some dead switch branches in old models (#28669)

2026-09-10 16:09:35 +03:00

nomic-bert.cpp

models: clean up some dead switch branches in old models (#28669)

2026-09-10 16:09:35 +03:00

olmo2.cpp

Model-Saver: Write the SWA pattern, 15 more architectures roundtrip (#29042)

2026-09-18 15:20:03 +02:00

olmo.cpp

hparams : refactor hparams.n_layer (#24060)

2026-06-05 11:09:36 +03:00

olmoe.cpp

convert : add --fuse-qkv flag to fuse Q/K/V into QKV during HF-to-GGUF conversion (#22780)

2026-09-07 00:47:05 +08:00

openai-moe.cpp

Model-Saver: Write the SWA pattern, 15 more architectures roundtrip (#29042)

2026-09-18 15:20:03 +02:00

openelm.cpp

hparams : refactor hparams.n_layer (#24060)

2026-06-05 11:09:36 +03:00

orion.cpp

hparams : refactor hparams.n_layer (#24060)

2026-06-05 11:09:36 +03:00

paddleocr.cpp

model : NvFP4 quantized LM head support (#23046)

2026-05-16 11:09:27 +02:00

pangu-embed.cpp

hparams : refactor hparams.n_layer (#24060)

2026-06-05 11:09:36 +03:00

phi2.cpp

hparams : refactor hparams.n_layer (#24060)

2026-06-05 11:09:36 +03:00

phi3.cpp

hparams : refactor hparams.n_layer (#24060)

2026-06-05 11:09:36 +03:00

phimoe.cpp

hparams : refactor hparams.n_layer (#24060)

2026-06-05 11:09:36 +03:00

plamo2.cpp

metal : single-source fusion table + fusion debug rework (#28164)

2026-09-11 12:41:54 +03:00

plamo3.cpp

Model-Saver: Write the SWA pattern, 15 more architectures roundtrip (#29042)

2026-09-18 15:20:03 +02:00

plamo.cpp

hparams : refactor hparams.n_layer (#24060)

2026-06-05 11:09:36 +03:00

plm.cpp

model: use ggml_rope_set_offset() (#27382)

2026-08-21 18:54:29 +02:00

pockettts.cpp

mtmd: support pocket-tts (#26871)

2026-08-11 14:18:30 +02:00

qwen2.cpp

hparams : refactor hparams.n_layer (#24060)

2026-06-05 11:09:36 +03:00

qwen2moe.cpp

model: add NVIDIA Nemotron-3-Puzzle-75B-A9B (NemotronHPuzzle) support (#25444)

2026-09-03 08:53:08 +02:00

qwen2vl.cpp

model : NvFP4 quantized LM head support (#23046)

2026-05-16 11:09:27 +02:00

qwen3.cpp

spec: add EAGLE3 speculative decoding support (#18039)

2026-06-12 10:21:06 +03:00

qwen3moe.cpp

model: add NVIDIA Nemotron-3-Puzzle-75B-A9B (NemotronHPuzzle) support (#25444)

2026-09-03 08:53:08 +02:00

qwen3next.cpp

convert : add --fuse-qkv flag to fuse Q/K/V into QKV during HF-to-GGUF conversion (#22780)

2026-09-07 00:47:05 +08:00

qwen3tts.cpp

mtmd: support Qwen3-TTS (note: breaking change to llama-tts binary) (#26254)

2026-08-04 17:26:15 +02:00

qwen3vl.cpp

metal : single-source fusion table + fusion debug rework (#28164)

2026-09-11 12:41:54 +03:00

qwen3vlmoe.cpp

model: add NVIDIA Nemotron-3-Puzzle-75B-A9B (NemotronHPuzzle) support (#25444)

2026-09-03 08:53:08 +02:00

qwen4exp.cpp

CUDA: enable sparse fa for qwen4 (#28770)

2026-09-20 16:08:11 +08:00

qwen35.cpp

convert : add --fuse-qkv flag to fuse Q/K/V into QKV during HF-to-GGUF conversion (#22780)

2026-09-07 00:47:05 +08:00

qwen35moe.cpp

convert : add --fuse-qkv flag to fuse Q/K/V into QKV during HF-to-GGUF conversion (#22780)

2026-09-07 00:47:05 +08:00

qwen.cpp

hparams : refactor hparams.n_layer (#24060)

2026-06-05 11:09:36 +03:00

refact.cpp

hparams : refactor hparams.n_layer (#24060)

2026-06-05 11:09:36 +03:00

rnd1.cpp

model: add NVIDIA Nemotron-3-Puzzle-75B-A9B (NemotronHPuzzle) support (#25444)

2026-09-03 08:53:08 +02:00

rwkv6-base.cpp

models : deduplicate delta-net graphs for Qwen family (#19597)

2026-02-16 14:35:04 +02:00

rwkv6.cpp

hparams : refactor hparams.n_layer (#24060)

2026-06-05 11:09:36 +03:00

rwkv6qwen2.cpp

hparams : refactor hparams.n_layer (#24060)

2026-06-05 11:09:36 +03:00

rwkv7-base.cpp

models : deduplicate delta-net graphs for Qwen family (#19597)

2026-02-16 14:35:04 +02:00

rwkv7.cpp

hparams : refactor hparams.n_layer (#24060)

2026-06-05 11:09:36 +03:00

seed-oss.cpp

hparams : refactor hparams.n_layer (#24060)

2026-06-05 11:09:36 +03:00

smallthinker.cpp

Model-Saver: Write the SWA pattern, 15 more architectures roundtrip (#29042)

2026-09-18 15:20:03 +02:00

smollm3.cpp

hparams : refactor hparams.n_layer (#24060)

2026-06-05 11:09:36 +03:00

spark2-5.cpp

[Model] Support for Spark2_5ForCausalLM implementation (#27868)

2026-09-06 17:43:58 +02:00

stablelm.cpp

hparams : refactor hparams.n_layer (#24060)

2026-06-05 11:09:36 +03:00

starcoder2.cpp

hparams : refactor hparams.n_layer (#24060)

2026-06-05 11:09:36 +03:00

starcoder.cpp

hparams : refactor hparams.n_layer (#24060)

2026-06-05 11:09:36 +03:00

step35.cpp

models : fix incorrect uses of get_key_or_arr (#28868)

2026-09-14 14:05:37 +03:00

t5.cpp

models : move build_arch_graph() after graph() template specialization (#28934)

2026-09-15 12:33:26 +03:00

t5encoder.cpp

model: move load_hparams and load_tensors to per-model definition (#22004)

2026-05-04 12:36:59 +02:00

talkie.cpp

hparams : refactor hparams.n_layer (#24060)

2026-06-05 11:09:36 +03:00

wavtokenizer-dec.cpp

model : NvFP4 quantized LM head support (#23046)

2026-05-16 11:09:27 +02:00

xverse.cpp

hparams : refactor hparams.n_layer (#24060)

2026-06-05 11:09:36 +03:00