ggerganov/llama.cpp

Pull Requests Commits
CUDA: fix RoPE asserts, block sizes (#2833)
JohannesGaessler committed 2 years ago
Verified 92b1bbd2
llama.h : add missing struct keyword for C compat in callback type (#2847)
igarnier committed 2 years ago
Verified dd0dc366
metal : fix memory leak (#2762)
ggerganov committed 2 years ago
Verified f55538c3
quantize : make output filename optional again (#2823)
cebtenzzre committed 2 years ago
Verified ebcee207
devops : added systemd units and set versioning to use date. (#2835)
jboero committed 2 years ago
Verified 3e8ff47a
gguf : fix strings to not be null-terminated (#2839)
ggerganov committed 2 years ago
Verified 103cfafc
llama : fix MPI threads (close #2827)
ggerganov committed 2 years ago
Verified c10704d0
examples : update llama2.c converter to read vocab and write models in GGUF format (#2751)
ochafik committed 2 years ago
Verified 230d46c7
llama : speedup tokenization (#2831)
ikawrakow committed 2 years ago
Verified 463173a6
falcon : fix CUDA inference by making K and Q contiguous (#2830)
ggerganov committed 2 years ago
Verified eaa13a48
readme : fix headings
ggerganov committed 2 years ago
Verified da7455d0
scripts : helper convert script
ggerganov committed 2 years ago
Verified 25423e91
k_quants tuning for Falcon-7b (#2816)
ikawrakow committed 2 years ago
Verified a6d1189f
readme : update hot topics
ggerganov committed 2 years ago
Verified c48c5bb0
gguf : add 64-bit support (GGUF v2) (#2821)
ggerganov committed 2 years ago
Verified d0cee0d3
llama : more tokenizer fixes (#2810)
ggerganov committed 2 years ago
Verified edd4c148
ggml : detect SSSE3 (#2825)
przemoc committed 2 years ago
Verified 1591e2e5
ci : add LoRA test to CI (#2650)
slaren committed 2 years ago
Verified 789c8c94
server : add `/detokenize` endpoint (#2802)
BruceMacD committed 2 years ago
Verified c1ac54b7
convert.py : advanced option (#2753)
KerfuffleV2 committed 2 years ago
Verified 730d9c68
llama : use Unicode Escape Sequence to replace encoded characters (#2814)
drasticactions committed 2 years ago
Verified c7d92e6d
flake.nix : add rocm support and cleanup (#2808)
Tungsten842 committed 2 years ago
Verified 61d1a289
llama : move #includes out of _GNU_SOURCE conditional (#2817)
cebtenzzre committed 2 years ago
Verified 741ca7dd
main : fix bug (penalize_nl=false doesn't work) + suppress warning on mingw (#1528)
tom7 committed 2 years ago
Verified 72f895c9
llama : use std::abs in llama_sample_tail_free (#2800)
cebtenzzre committed 2 years ago
Verified 50526f37
k-quants : remove unnecessary tensor shape restrictions (#2811)
ggerganov committed 2 years ago
Verified 04f4b1eb
Better perplexity for 2- and 3-bit quantization for LLaMA-v2-70B (#2807)
ikawrakow committed 2 years ago
Verified 75923754
Fix HellaSwag (#2805)
ikawrakow committed 2 years ago
Verified 771551a7
flake : build llama.cpp on Intel with nix (#2795)
signalpillar committed 2 years ago
Verified f305bad1
Handle null rope scaling value (#2793)
pnb committed 2 years ago
Verified a2ca4e9d
Terms Of Use Privacy Policy Refund Policy Impressum