ngxson/llama.cpp

Pull Requests Commits

Merge branch 'master' into xsn/fix_lora

ngxson committed 1 year ago

383b6bce

ngxson committed 1 year ago

Verified 4db8f60f

convert_hf : move add_type to main()

compilade committed 1 year ago

d09382fa

Merge branch 'master' into xsn/fix_lora

ngxson committed 1 year ago

4d9ac0f3

move add_type to __init__

ngxson committed 1 year ago

b1c40695

change kv metadata

ngxson committed 1 year ago

0ba23bad

Apply suggestions from code review

ngxson committed 1 year ago

Verified 9175f4b7

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

Merge branch 'master' into xsn/fix_lora

ngxson committed 1 year ago

b704448a

llama : de-duplicate deepseek2 norm

ggerganov committed 1 year ago

Verified 3dfda059

remove redundant params

ngxson committed 1 year ago

f68d0924

Revert "auto scale"

ngxson committed 1 year ago

5b181182

ngxson committed 1 year ago

42415a48

Merge branch 'master' into xsn/fix_lora

ngxson committed 1 year ago

703573f6

Vulkan MMQ Fix (#8479)

0cc4m committed 1 year ago

Verified bda62d79

llama : use llm_build_lora_mm in most model graphs

compilade committed 1 year ago

87301bdd

convert_hf : simplify modify_tensors for InternLM2

compilade committed 1 year ago

8956543c

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

compilade committed 1 year ago

Verified 090fca7a

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

Older