History for examples/low_level_api - abetlen/llama-cpp-python · GitHub

github.com

Tin mới

feat: update llama.cpp to 936918514 (#2348)

fix(misc): replace deprecated llama.cpp references (#2170)

examples: fix quantize example (#1387)

Fix low_level_api_chat_cpp example to match current API (#1086)

Update low_level_api_llama_cpp.py to match current API (#1023)

Add numa support, low level api users must now explicitly call llama_backend_init at the start of their programs.

Fix low level api examples

Merge pull request #265 from dmahurin/fix-from-bytes-byteorder

Merge pull request #274 from dmahurin/fix-missing-antiprompt

low_level_api_chat_cpp.py: Fix missing antiprompt output in chat.

fix "missing 1 required positional argument: 'min_keep'"

fix "from_bytes() missing required argument 'byteorder'"

Fix session loading and saving in low level example chat

Fix a bug with wrong type

Add utf8 to chat example

Fix mirastat requiring c_float

Wrong logit_bias parsed type

Rename postfix to suffix to match upstream

Update low level examples

Detect multi-byte responses and wait

Also ignore errors on input prompts

Remove try catch from chat

Fix decode errors permanently

Fixed end of text wrong type, and fix n_predict behaviour

More reasonable defaults

Added iterative search to prevent instructions from being echoed, add ignore eos, add no-mmap, fixed 1 character echo too much bug

Update model paths to be more clear they should point to file

More interoperability to the original llama.cpp, and arguments now work

Fixed too many newlines, now onto args.

Better llama.cpp interoperability

Fix bug in init_break not being set when exited via antiprompt and others.

Move to new examples directory