ngxson/llama.cpp

Pull Requests Commits

ngxson committed 238 days ago

5c039a72

ngxson committed 238 days ago

3894eb95

update llava/readme

ngxson committed 238 days ago

d7e95ae8

ah sheet it works

ngxson committed 238 days ago

c53ecc96

load ok, missing patch merger

ngxson committed 238 days ago

efce8750

ngxson committed 239 days ago

656759ae

arg : -hf do not fail if url mismatch (#13219)

ngxson committed 239 days ago

Verified 6f67cf1f

fix typo: `n_ctx_pre_seq` -> `n_ctx_per_seq` (#13221)

ddh0 committed 239 days ago

Verified 16a457fa

convert : improve model arch handling (#13122)

ngxson committed 239 days ago

Verified 3e168bed

llava : remove duplicate include (#13207)

tattn committed 239 days ago

Verified ceda28ef

common : add -jf / --json-schema-file flag (#12011)

ochafik committed 239 days ago

Verified 3b127c73

vulkan: use uint array index to avoid glslang bug (#13193)

jeffbolznv committed 239 days ago

Verified e5007a5e

ggml : fix ppc64le build (#13176)

shalinib-ibm committed 239 days ago

Verified 41631377

convert : correct typo image_mean --> image_std (#13208)

ngxson committed 239 days ago

Verified 07c2e2f7

feat(ggml-cpu): enable z17 compile (#13182)

taronaeo committed 239 days ago

Verified 44cd8d91

arg : allow using -hf offline (#13202)

ngxson committed 239 days ago

Verified 5933e6fd

docker : do not build tests (#13204)

ngxson committed 239 days ago

Verified da84c04d

rpc : fix cache directory initialization (#13188)

hbuxiaofei committed 239 days ago

Verified a0f7016d

scripts: n_depth for compare-llama-bench [no ci] (#13201)

JohannesGaessler committed 240 days ago

Verified 19e899ce

server : Prefilling assistant message in openai compatible API (#13174)

matteoserva committed 240 days ago

Verified e2e1ddb9

sampling : when top-k <= 0 -> noop (#13173)

ggerganov committed 240 days ago

Verified d9d398f8

llama-bench: fixed size of fields to correctly map to values (#13183)

Alberto Cabrera Pérez committed 240 days ago

Verified 5a639801

CUDA: fix non-cont. inputs for batched mat mul (#13155)

JohannesGaessler committed 240 days ago

Verified cdf76586

llama : llm_type order by size (#13177)

CISC committed 240 days ago

Verified 7d3af70b

mtmd : add qwen2vl and qwen2.5vl (#13141)

ngxson committed 240 days ago

Verified 00e3e5a1

llama : set qwen3 model type sizes (#13175)

CISC committed 240 days ago

Verified e98b3692

llama-graph : fix text position for mrope (#13159)

ngxson committed 240 days ago

Verified b6ce7430

model : Nomic Embed Text V2 with Mixture-of-Experts (MoE) architecture (#12466)

manyoso committed 241 days ago

Verified 5f5e39e1

clip : fix model size display (#13153)

ngxson committed 241 days ago

Verified eaea3253

fix(rpc): Improve input validation and error handling (#13069)

thevilledev committed 241 days ago

Verified 43ddab6e

Older