ggerganov/llama.cpp

Pull Requests Commits

convert_hf : fix memory leak in lazy MoE conversion

compilade committed 1 year ago

b971122e

Merge branch 'master' into compilade/faster-lazy-safetensors

compilade committed 1 year ago

2a49a68d

Refactor lora adapter support (#8332)

ngxson committed 1 year ago

Verified 97bdd26e

ngxson committed 1 year ago

Verified 4db8f60f

ggml : suppress unknown pragma 'GCC' on windows (#8460)

danbev committed 1 year ago

Verified 8fac431b

server: update README.md with llama-server --help output [no ci] (#8472)

maruel committed 1 year ago

Verified f17f39ff

common : add --no-cont-batching arg (#6358)

ggerganov committed 1 year ago

Verified 9104bc20

docs: fix links in development docs [no ci] (#8481)

NikolaiLyssogor committed 1 year ago

Verified fc690b01

[SYCL] add concat through dim 1/2 (#8483)

airMeng committed 1 year ago

Verified 16bdfa42

llama : de-duplicate deepseek2 norm

ggerganov committed 1 year ago

Verified 3dfda059

Vulkan MMQ Fix (#8479)

0cc4m committed 1 year ago

Verified bda62d79

pydantic : replace uses of __annotations__ with get_type_hints (#8474)

compilade committed 1 year ago

Verified 090fca7a

convert_hf : faster lazy safetensors

compilade committed 1 year ago

7cda4dd7

flake.lock: Update (#8475)

ggerganov committed 1 year ago

Verified aaab2419

llama : fix Gemma-2 Query scaling factors (#8473)

ggerganov committed 1 year ago

Verified 73cf442e

gguf_hash.py: Add sha256 (#8470)

mofosyne committed 1 year ago

Verified e236528e

llama : fix pre-tokenization of non-special added tokens (#8228)

compilade committed 1 year ago

Verified fa79495b

vulkan : cmake integration (#8119)

bandoti committed 1 year ago

Verified 17eb6aa8

metal : template-ify some of the kernels (#8447)

ggerganov committed 1 year ago

Verified c917b67f

server : handle content array in chat API (#8449)

ggerganov committed 1 year ago

Verified 4e24cffd

main : print error on empty input (#8456)

ggerganov committed 1 year ago

Verified 6af51c0d

llama : suppress unary minus operator warning (#8448)

danbev committed 1 year ago

Verified f5322624

server : ensure batches are either all embed or all completion (#8420)

iamlemec committed 1 year ago

Verified c3ebcfa1

docker : fix filename for convert-hf-to-gguf.py in tools.sh (#8441)

kriation committed 1 year ago

Verified 8a4441ea

convert : remove fsep token from GPTRefactForCausalLM (#8237)

jpodivin committed 1 year ago

Verified 5aefbce2

examples : sprintf -> snprintf (#8434)

ggerganov committed 1 year ago

Verified 71c1121d

ggml : minor naming changes (#8433)

ggerganov committed 1 year ago

Verified 370b1f7e

[SYCL] fix the mul_mat_id ut issues (#8427)

ClarkChin08 committed 1 year ago

Verified b549a1bb

ggml : add NVPL BLAS support (#8329) (#8425)

nicholaiTukanov committed 1 year ago

Verified 36864569

cuda : suppress 'noreturn' warn in no_device_code (#8414)

danbev committed 1 year ago

Verified b078c619

Older