ngxson/llama.cpp

Pull Requests Commits

Merge branch 'master' into xsn/cleanup_oai

ngxson committed 2 years ago

51f80eb7

server : add SSL support (#5926)

gabe-l-hart committed 2 years ago

Verified e1fa9569

server: tests: add truncated prompt tests, better kv cache size (#5933)

phymbert committed 2 years ago

Verified fd72d2d2

llama : support Mamba Selective State Space Models (#5328)

compilade committed 2 years ago

Verified c2101a2e

ngxson committed 2 years ago

09d3b658

fix test case CORS Options

ngxson committed 2 years ago

1a7c5fd5

llama : fix quantization of shared token_embd (#5944)

compilade committed 2 years ago

Verified 515f7d0d

fix embedding response

ngxson committed 2 years ago

7445bb8a

correct http verb for endpoints

ngxson committed 2 years ago

bd525411

Merge branch 'master' into xsn/cleanup_oai

ngxson committed 2 years ago

a1351efc

merge embedding handlers

ngxson committed 2 years ago

1866e185

use set_pre_routing_handler for validate_api_key

ngxson committed 2 years ago

07f120eb

server: metrics: add llamacpp:prompt_seconds_total and llamacpp:tokens_predicted_seconds_total, reset bucket only on /metrics. Fix values cast to int. Add Process-Start-Time-Unix header. (#5937)

phymbert committed 2 years ago

Verified 76e86882

refactor static file handler

ngxson committed 2 years ago

d8a8bd4c

llama : assume tied weights if lm_head/output weights is missing (#5824)

dmahurin committed 2 years ago

Verified e457fb35

server : fix EOS token detection with disabled cache (#5938)

ggerganov committed 2 years ago

Verified af37fd8b

log : fix MSVC compile errors (#5643)

uextm committed 2 years ago

Verified 581ed5c4

llama-bench : add embeddings option (#5924)

ggerganov committed 2 years ago

Verified 6cdabe65

Revert "[SYCL] fix error when set main gpu to non-zero (#5901)" (#5918)

NeoZhangJianyu committed 2 years ago

Verified 89fb735f

server : add `/v1/completions` endpoint (#5914)

mscheong01 committed 2 years ago

Verified 55a2a900

server : refactor (#5882)

ggerganov committed 2 years ago

Verified 2002bc96

[SYCL] fix error when set main gpu to non-zero (#5901)

NeoZhangJianyu committed 2 years ago

Verified ceca1aef

ggml : use SYS_get_cpu if SYS_getcpu is not defined (#5906)

cebtenzzre committed 2 years ago

Verified e04e04f8

ggml : use `uint8x16_t` return type for `ggml_vqtbl1q_u8` (#5894)

bobqianic committed 2 years ago

Verified e25fb4b1

convert : remove AWQ remnants (#5768)

ggerganov committed 2 years ago

Verified 1e35d619

add wait() to make code stable (#5895)

NeoZhangJianyu committed 2 years ago

Verified 8ced9f7e

compare-llama-bench.py : remove mul_mat_q (#5892)

slaren committed 2 years ago

Verified 652ca2bd

quants : use MM256_SET_M128I consistently to fix gcc 7 build (#5889)

cebtenzzre committed 2 years ago

Verified bd836944

grammars : blacklists character control set (#5888)

ExtReMLapin committed 2 years ago

Verified 3de31677

Revert "grammars : don't allow to output unescaped new line in string (#5885)"

ggerganov committed 2 years ago

Verified 82cb31eb

Older