Georgi Gerganov
|
afd220d9c6
Properly free llama_context on failure
|
пре 2 година |
Cameron Kaiser
|
481044d50c
additional optimizations for POWER9 (#454)
|
пре 2 година |
comex
|
563cdc391d
Support calling mlock() on loaded model data on Linux and macOS (#453)
|
пре 2 година |
Luciano
|
8d4a855c24
Add embedding mode with arg flag. Currently working (#282)
|
пре 2 година |
Georgi Gerganov
|
b6b268d441
Add link to Roadmap discussion
|
пре 2 година |
Georgi Gerganov
|
3cd8dde0d1
Revert "Fix memory allocation issues and seg faults"
|
пре 2 година |
Georgi Gerganov
|
4870e455b3
Fix memory allocation issues and seg faults
|
пре 2 година |
Georgi Gerganov
|
483bab2e3d
Avoid the transposed X branch in the Z = X * Y matrix multiplication (#439)
|
пре 2 година |
Jed Fox
|
404e1da38e
Fix quantize script not finding models in parent directory (#428)
|
пре 2 година |
Georgi Gerganov
|
4cc053b6d5
Remove oboslete command from Docker script
|
пре 2 година |
Georgi Gerganov
|
0ba5a3a9a5
Obsolete
|
пре 2 година |
rabidcopy
|
2e17dfd80a
Replace EOS with newline to prevent context/memory being flushed by EOS in interactive mode (#333)
|
пре 2 година |
Timmy Knight
|
20a1a4e09c
Fix GPTQ converter (#423)
|
пре 2 година |
nusu-github
|
ad072fc5ad
Generate library with CMake (#430)
|
пре 2 година |
anzz1
|
ea10d3ded2
Command line args bounds checking (#424)
|
пре 2 година |
Ben Siraphob
|
a18c19259a
Fix Nix build
|
пре 2 година |
Stephan Walter
|
a50e39c6fe
Revert "Delete SHA256SUMS for now" (#429)
|
пре 2 година |
Kerfuffle
|
a140219e81
Fix Makefile echo escape codes (by removing them). (#418)
|
пре 2 година |
Gary Mulder
|
8a3e5ef801
Move model section from issue template to README.md (#421)
|
пре 2 година |
anzz1
|
8eea5ae0e5
Delete SHA256SUMS for now (#416)
|
пре 2 година |
Georgi Gerganov
|
93208cfb92
Adjust repetition penalty ..
|
пре 2 година |
Georgi Gerganov
|
03ace14cfd
Add link to recent podcast about whisper.cpp and llama.cpp
|
пре 2 година |
anzz1
|
e4412b45e3
CI: CMake: Separate build and test steps (#376)
|
пре 2 година |
tjohnman
|
f7dc43bc0d
Fix instruct mode broken by PR #354 (#409)
|
пре 2 година |
Gary Mulder
|
ee8a788786
Update issue template so people will use it (#404)
|
пре 2 година |
Stephan Walter
|
69c92298a9
Deduplicate q4 quantization functions (#383)
|
пре 2 година |
Valentyn Bezshapkin
|
97940520e8
fix: add POSIX functionality for Linux compilation (#51)
|
пре 2 година |
tjohnman
|
305ba6f0e6
Don't force immediate interactive without `-i` (#354)
|
пре 2 година |
Erik Scholz
|
4122dffff9
cmake: make llama an actual library (#392)
|
пре 2 година |
Erik Scholz
|
56e659a0b2
fix perplexity after c-api refactor (#390)
|
пре 2 година |