Commits

Branch selector

 main

User selector

All users

Datepicker

All time

Commit History

Commits on May 10, 2024

• docs: Fix typo in README.md (#1444)

  yupbank

  authored
  1547202

  Copy full SHA for 1547202

• fix: Enable CUDA backend for llava. Closes #1324

  abetlen

  committed
  7f59856

  Copy full SHA for 7f59856

• chore: Bump version

  abetlen

  committed
  7316502

  Copy full SHA for 7316502

• feat: Update llama.cpp

  abetlen

  committed
  eafb6ec

  Copy full SHA for eafb6ec

• fix: Clear kv cache to avoid kv bug when image is evaluated first

  abetlen

  committed
  ac55d0a

  Copy full SHA for ac55d0a

• chore: Bump version

  abetlen

  committed
  4badac3

  Copy full SHA for 4badac3

• fix(security): Render all jinja templates in immutable sandbox (#1441)

  Show description for 561e880
  

  CISC

  and

  abetlen

  authored
  561e880

  Copy full SHA for 561e880

• Merge pull request from GHSA-56xg-wfcc-g829

  Show description for b454f40
  

  retr0reg

  and

  abetlen

  authored
  b454f40

  Copy full SHA for b454f40

Commits on May 9, 2024

• feat: Support multiple chat templates - step 1 (#1396)

  Show description for 5ab40e6
  

  CISC

  and

  abetlen

  authored
  5ab40e6

  Copy full SHA for 5ab40e6

• chore: Bump version

  abetlen

  committed
  bf66a28

  Copy full SHA for bf66a28

• fix: free last image embed in llava chat handler

  abetlen

  committed
  3757328

  Copy full SHA for 3757328

Commits on May 8, 2024

• fix: Make leading bos_token optional for image chat formats, fix nanollava system message

  abetlen

  committed
  7712263

  Copy full SHA for 7712263

• feat: Update llama.cpp

  abetlen

  committed
  2a39b99

  Copy full SHA for 2a39b99

• chore: Bump version

  abetlen

  committed
  9ce5cb3

  Copy full SHA for 9ce5cb3

• feat: fill-in-middle support (#1386)

  Show description for 4a7122d
  

  CISC

  and

  abetlen

  authored
  4a7122d

  Copy full SHA for 4a7122d

• feat: Update llama.cpp

  abetlen

  committed
  228949c

  Copy full SHA for 228949c

• fix: adding missing args in create_completion for functionary chat handler (#1430)

  skalade

  authored
  903b28a

  Copy full SHA for 903b28a

• docs: update README.md (#1432)

  Show description for 07966b9
  

  Ikko Eltociear Ashimine

  authored
  07966b9

  Copy full SHA for 07966b9

• fix: chat_format log where auto-detected format prints `None` (#1434)

  balvisio

  authored
  a50d24e

  Copy full SHA for a50d24e

Commits on May 5, 2024

• feat(server): Add support for setting root_path. Closes #1420

  abetlen

  committed
  0318702

  Copy full SHA for 0318702

• feat(ci): Add docker checks and check deps more frequently (#1426)

  Show description for 3666833
  Smartappli

  authored
  3666833

  Copy full SHA for 3666833

• feat: Update llama.cpp

  abetlen

  committed
  3e2597e

  Copy full SHA for 3e2597e

Commits on May 4, 2024

• fix: detokenization case where first token does not start with a leading space (#1375)

  Show description for e0d7674
  

  noamgat

  and

  abetlen

  authored
  e0d7674

  Copy full SHA for e0d7674

• feat: Implement streaming for Functionary v2 + Bug fixes (#1419)

  Show description for 1f56c64
  

  jeffrey-fong

  and

  abetlen

  authored
  1f56c64

  Copy full SHA for 1f56c64

Commits on May 3, 2024

• Merge branch 'main' of github.com:abetlen/llama_cpp_python into main

  abetlen

  committed
  f9b7221

  Copy full SHA for f9b7221

• fix: Use memmove to copy str_value kv_override. Closes #1417

  abetlen

  committed
  9f7a855

  Copy full SHA for 9f7a855

• feat(server): Remove temperature bounds checks for server. Closes #1384

  abetlen

  committed
  0a454be

  Copy full SHA for 0a454be

• fix(server): Propagate `flash_attn` to model load. (#1424)

  dthuerck

  authored
  2138561

  Copy full SHA for 2138561

Commits on May 2, 2024

• chore: Bump version

  abetlen

  committed
  2117122

  Copy full SHA for 2117122

• feat: Update llama.cpp

  abetlen

  committed
  d75dea1

  Copy full SHA for d75dea1

• feat: Add llama-3-vision-alpha chat format

  abetlen

  committed
  31b1d95

  Copy full SHA for 31b1d95

Commits on Apr 30, 2024

• fix: Change default verbose value of verbose in image chat format handlers to True to match Llama

  abetlen

  committed
  4f01c45

  Copy full SHA for 4f01c45

• feat: Update llama.cpp

  abetlen

  committed
  946156f

  Copy full SHA for 946156f

• Merge branch 'main' of github.com:abetlen/llama_cpp_python into main

  abetlen

  committed
  9286b5c

  Copy full SHA for 9286b5c

• fix: Suppress all logs when verbose=False, use hardcoded fileno's to work in colab notebooks. Closes #796 Closes #729

  abetlen

  committed
  f116175

  Copy full SHA for f116175

Pagination

Previous

Next