Commits

Branch selector

 main

User selector

All users

Datepicker

All time

Commit History

Commits on May 20, 2023

• Fixed logit processors

  Maximilian-Winter

  committed
  e2a9599

  Copy full SHA for e2a9599

• Merge branch 'main' of https://github.com/Maximilian-Winter/llama-cpp-python

  Maximilian-Winter

  committed
  fb0e9a5

  Copy full SHA for fb0e9a5

• Fix when stopping criteria's are none

  Maximilian-Winter

  committed
  00dd8ac

  Copy full SHA for 00dd8ac

• Merge branch 'abetlen:main' into main

  Maximilian-Winter

  authored
  6145d54

  Copy full SHA for 6145d54

• Bump version

  abetlen

  committed
  8f49ca0

  Copy full SHA for 8f49ca0

• Change properties to functions to match token functions

  abetlen

  committed
  76b1d2c

  Copy full SHA for 76b1d2c

• Merge branch 'abetlen:main' into main

  Maximilian-Winter

  authored
  8fe0c3c

  Copy full SHA for 8fe0c3c

• Added support for logic processors and dynamic stop criteria's

  Maximilian-Winter

  committed
  33629b4

  Copy full SHA for 33629b4

• Add n_ctx, n_vocab, and n_embd properties

  abetlen

  committed
  a7ba858

  Copy full SHA for a7ba858

Commits on May 19, 2023

• Fix llama_cpp and Llama type signatures. Closes #221

  abetlen

  committed
  01a010b

  Copy full SHA for 01a010b

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

  abetlen

  committed
  fb57b94

  Copy full SHA for fb57b94

• Bump version

  abetlen

  committed
  f82d85f

  Copy full SHA for f82d85f

• Add Guidance example

  abetlen

  committed
  c7788c8

  Copy full SHA for c7788c8

• Bugfix: Stop sequences can be strings

  abetlen

  committed
  a8cd169

  Copy full SHA for a8cd169

• Add upgrade instructions to the README

  abetlen

  committed
  f0812c4

  Copy full SHA for f0812c4

• Fix logprobs for completions and implement for streaming logprobs.

  abetlen

  committed
  17d4271

  Copy full SHA for 17d4271

• Allow first logprob token to be null to match openai api

  abetlen

  committed
  a634a24

  Copy full SHA for a634a24

• Use server sent events function for streaming completion

  abetlen

  committed
  dc39cc0

  Copy full SHA for dc39cc0

Commits on May 18, 2023

• Merge pull request #235 from Pipboyguy/main

  Show description for 69f9d50
  abetlen

  authored
  69f9d50

  Copy full SHA for 69f9d50

• Stream tokens instead of text chunks

  abetlen

  committed
  f0ec6e6

  Copy full SHA for f0ec6e6

• Remove unnused union

  abetlen

  committed
  21d8f5f

  Copy full SHA for 21d8f5f

• Set CUDA_VERSION as build ARG

  Show description for 6ece8a2
  Pipboyguy

  committed
  6ece8a2

  Copy full SHA for 6ece8a2

• Decrement CUDA version and bump Ubuntu

  Show description for 6c57d38
  Pipboyguy

  committed
  6c57d38

  Copy full SHA for 6c57d38

Commits on May 17, 2023

• Update llama.cpp

  abetlen

  committed
  50e1362

  Copy full SHA for 50e1362

• Update docs

  abetlen

  committed
  db10e00

  Copy full SHA for db10e00

• Check for CUDA_PATH before adding

  abetlen

  committed
  61d58e7

  Copy full SHA for 61d58e7

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

  abetlen

  committed
  7c95895

  Copy full SHA for 7c95895

• Merge pull request #225 from aneeshjoy/main

  Show description for 47921a3
  abetlen

  authored
  47921a3

  Copy full SHA for 47921a3

• Fixd CUBLAS dll load issue in Windows

  aneeshjoy

  committed
  e9794f9

  Copy full SHA for e9794f9

• Move docs link up

  abetlen

  committed
  70695c4

  Copy full SHA for 70695c4

• Update token checks

  abetlen

  committed
  4f34279

  Copy full SHA for 4f34279

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

  abetlen

  committed
  626003c

  Copy full SHA for 626003c

• Format

  abetlen

  committed
  f5c2f99

  Copy full SHA for f5c2f99

• Implement penalize_nl

  abetlen

  committed
  d28b753

  Copy full SHA for d28b753

• Fix last_n_tokens_size

  abetlen

  committed
  f11e2a7

  Copy full SHA for f11e2a7

Pagination

Previous

Next