ngxson/llama.cpp

Pull Requests Commits

vocab: fix Gemma4 tokenizer (#21343)

pwilkin committed 65 days ago

Verified b069b10a

rpc : reuse compute graph buffers (#21299)

rgerganov committed 65 days ago

Verified 0c58ba33

chat : avoid including json in chat.h (#21306)

ggerganov committed 65 days ago

Verified 57ace0d6

(revert) kv-cache : do not quantize SWA KV cache (#21332)

ggerganov committed 65 days ago

Verified 39b27f0d

ci : add AMD ZenDNN label to PR labeler (#21345)

z-vishal committed 65 days ago

Verified f49e9178

[HIP] Bump ROCm version to 7.2.1 (#21066)

slojosic-amd committed 66 days ago

Verified 7c7d6ce5

fix: gemma 4 template (#21326)

pwilkin committed 66 days ago

Verified 5208e2d5

tests : add unit test coverage for llama_tensor_get_type (#20112)

bartowski1182 committed 66 days ago

Verified 7992aa7c

ggml-webgpu: add vectorized flash attention (#20709)

ArberSephirotheca committed 66 days ago

Verified a1cfb645

tests: allow exporting graph ops from HF file without downloading weights (#21182)

0cc4m committed 66 days ago

Verified 5803c8d1

model, mtmd: fix gguf conversion for audio/vision mmproj (#21309)

ngxson committed 66 days ago

Verified 63f8fe0e

common : add commentary rules for gpt-oss-20b (#21286)

aldehir committed 66 days ago

Verified 22337374

Relax prefill parser to allow space. (#21240)

pwilkin committed 66 days ago

Verified e15efe00

chat : add Granite 4.0 chat template with correct tool_call role mapping (#20804)

jesus-talavera-ibm committed 66 days ago

Verified 6137c325

kv-cache : do not quantize SWA KV cache (#21277)

ggerganov committed 66 days ago

Verified 17193cce

Ignore Transfer-Encoding header. (#20269)

crmky committed 66 days ago

Verified d6dac92b

ggerganov committed 66 days ago

dae2bf41

ggml : bump version to 0.9.11 (ggml/1456)

ggerganov committed 66 days ago

bc07d559

sycl : fix llama_kv_cache hang when kv_cache is huge: 5GB (#21283)

arthw committed 66 days ago

Verified 4888137b

hexagon : add cumsum op support (#21246)

tboinovski1 committed 66 days ago

Verified fbd441c3

contrib : rewrite AGENTS.md, make it more clear about project values (#21270)

ngxson committed 67 days ago

Verified c30e0122

opencl: fix leak in Adreno q8_0 path (#21212)

lhez committed 67 days ago

Verified 95a6ebab

server: Bypass API Key validation for WebUI static bundle assets (#21269)

allozaur committed 67 days ago

Verified 12dbf1da

CUDA: fix FA kernel selection logic (#21271)

JohannesGaessler committed 67 days ago

Verified 86221cf6

kleidiai: add CPU feature detection to CI run script (#20394)

martin-klacer-arm committed 67 days ago

Verified 6de97b9d

Update Dawn version in WebGPU CI (#20784)

nikhilJain17 committed 67 days ago

Verified 5a0ed515

hexagon: improve RMS_NORM and DIV accuracy (#21251)

aparmp-quic committed 67 days ago

Verified 8710e5f9

fix: tool call parsing for LFM2 and LFM2.5 models (#21242)

jbuchananr committed 67 days ago

Verified 1d6d4cf7

llama : rotate activations for better quantization (#21038)

ggerganov committed 67 days ago

Verified 744c0c73

scripts: add function call test script (#21234)

ngxson committed 67 days ago

Verified 0356e33a

Newer Older