Commits

Branch selector

 master

User selector

danbev

All time

Commit History

Commits on Feb 15, 2026

• cmake : check if KleidiAI API has been fetched (#19640)

  Show description for 5708827
  danbev

  authored
  5708827

  Copy full SHA for 5708827

Commits on Feb 12, 2026

• scripts : add support for forks in pr2wt.sh (#19540)

  Show description for ff59903
  danbev

  authored
  ff59903

  Copy full SHA for ff59903

Commits on Feb 11, 2026

• common : remove unused token util functions (#19506)

  Show description for 3136a84
  danbev

  authored
  3136a84

  Copy full SHA for 3136a84

• build : fix case in dSYMs path for build-macos [no ci] (#19515)

  Show description for 53de59f
  danbev

  authored
  53de59f

  Copy full SHA for 53de59f

• llama : refactor sampling_info to use buffer_view template (#19368)

  Show description for 2cce9fd
  danbev

  authored
  2cce9fd

  Copy full SHA for 2cce9fd

Commits on Feb 10, 2026

• tts : fix typos in README.md [no ci] (#19463)

  danbev

  authored
  66d403c

  Copy full SHA for 66d403c

Commits on Feb 6, 2026

• llama : rename llama-sampling to llama-sampler (#19363)

  Show description for e696cfc
  danbev

  authored
  e696cfc

  Copy full SHA for e696cfc

Commits on Feb 4, 2026

• codeowners : add danbev for examples/debug (#19332)

  Show description for b536eb0
  

  danbev

  and

  pwilkin

  authored
  b536eb0

  Copy full SHA for b536eb0

• model-conversion : add tensor-info.py utility (#18954)

  Show description for 6ab881b
  danbev

  authored
  6ab881b

  Copy full SHA for 6ab881b

• completion : simplify batch (embd) processing (#19286)

  Show description for 25f40ca
  danbev

  authored
  25f40ca

  Copy full SHA for 25f40ca

Commits on Feb 2, 2026

• model-conversion : add debug option to conversion script (#19265)

  Show description for 6156ae5
  danbev

  authored
  6156ae5

  Copy full SHA for 6156ae5

Commits on Jan 30, 2026

• memory : clarify comments for r_l and s_l tensors [no ci] (#19203)

  Show description for f3bc988
  danbev

  authored
  f3bc988

  Copy full SHA for f3bc988

• convert : add missing return statement for GraniteMoeModel (#19202)

  Show description for 0562503
  danbev

  authored
  0562503

  Copy full SHA for 0562503

• memory : remove unused tmp_buf (#19199)

  Show description for 83bcdf7
  danbev

  authored
  83bcdf7

  Copy full SHA for 83bcdf7

Commits on Jan 28, 2026

• convert : yield Mamba2Model/GraniteMoeModel modify_tensors (#19157)

  Show description for ebf5725
  danbev

  authored
  ebf5725

  Copy full SHA for ebf5725

• sampling : remove sampling branching in output_reserve (#18811)

  Show description for eef375c
  danbev

  authored
  eef375c

  Copy full SHA for eef375c

Commits on Jan 27, 2026

• common : clarify HTTPS build options in error message (#19103)

  Show description for fc3cdf3
  danbev

  authored
  fc3cdf3

  Copy full SHA for fc3cdf3

Commits on Jan 25, 2026

• common : use two decimal places for float arg help messages (#19048)

  Show description for 16639ba
  danbev

  authored
  16639ba

  Copy full SHA for 16639ba

Commits on Jan 23, 2026

• model-conversion : use BUILD_DIR variable in all scripts (#19015)

  Show description for a14b960
  danbev

  authored
  a14b960

  Copy full SHA for a14b960

Commits on Jan 21, 2026

• llama : clarify nemotron-h.cpp comment about RoPE [no ci] (#18997)

  Show description for 9da3dcd
  danbev

  authored
  9da3dcd

  Copy full SHA for 9da3dcd

Commits on Jan 20, 2026

• convert : use n_groups instead of hardcoded values in reshape (#18929)

  Show description for 7dee9ff
  danbev

  authored
  7dee9ff

  Copy full SHA for 7dee9ff

Commits on Jan 19, 2026

• model-conversion : add BUILD_DIR variable to run-converted-model scripts (#18927)

  Show description for 3d55846
  danbev

  authored
  3d55846

  Copy full SHA for 3d55846

Commits on Jan 14, 2026

• CUDA : fix typo in clang pragma comment [no ci] (#18830)

  danbev

  authored
  01cbdfd

  Copy full SHA for 01cbdfd

Commits on Jan 13, 2026

• model-conversion : remove -c 0 from model card template [no ci] (#18807)

  Show description for 20ca2e1
  danbev

  authored
  20ca2e1

  Copy full SHA for 20ca2e1

Commits on Jan 12, 2026

• examples : add --kv-unified to batched example (#18774)

  Show description for 4150da9
  danbev

  authored
  4150da9

  Copy full SHA for 4150da9

Commits on Jan 11, 2026

• debug : include LLAMA_POOLING_TYPE_UNSPECIFIED in pooling check (#18692)

  Show description for 9789e28
  danbev

  authored
  9789e28

  Copy full SHA for 9789e28

Commits on Jan 8, 2026

• model-conversion : add warn about transformers mismatch (#18691)

  Show description for 9c142e3
  danbev

  authored
  9c142e3

  Copy full SHA for 9c142e3

• model-conversion : remove -st targets for converted model (#18689)

  Show description for df7fb92
  danbev

  authored
  df7fb92

  Copy full SHA for df7fb92

Commits on Jan 7, 2026

• convert : clarify sentence-transformers-dense-modules help [no ci] (#18662)

  Show description for bb77764
  danbev

  authored
  bb77764

  Copy full SHA for bb77764

• examples : add debug utility/example (#18464)

  Show description for ffba4f2
  danbev

  authored
  ffba4f2

  Copy full SHA for ffba4f2

Commits on Jan 4, 2026

• sampling : add support for backend sampling (#17004)

  Show description for d3dce4e
  

  4 peopleauthored
  d3dce4e

  Copy full SHA for d3dce4e

Commits on Dec 30, 2025

• llama : fix typo in comment in llama-kv-cache.h [no ci] (#18489)

  danbev

  authored
  ac1d0eb

  Copy full SHA for ac1d0eb

• model-conversion : use CONVERTED_MODEL for compare-embeddings (#18461)

  Show description for a864fb1
  danbev

  authored
  a864fb1

  Copy full SHA for a864fb1

Commits on Dec 29, 2025

• model-conversion : add device option to embd run orig model (#18386)

  Show description for 7cbec34
  danbev

  authored
  7cbec34

  Copy full SHA for 7cbec34

Commits on Dec 23, 2025

• model-conversion : add device option to run-org-model.py (#18318)

  Show description for 8e3ead6
  danbev

  authored
  8e3ead6

  Copy full SHA for 8e3ead6