ngxson/llama.cpp

Pull Requests Commits

ggerganov committed 67 days ago

6422036f

ggml : bump version to 0.9.10 (ggml/1454)

ggerganov committed 67 days ago

296bc053

sycl : support nvfp4 type in mul_mat (#21227)

arthw committed 67 days ago

Verified 6b949d10

ggml-cuda: Add generic NVFP4 MMQ kernel (#21074)

michaelw9999 committed 67 days ago

Verified 84f82e84

memory: respect unified KV cache in hybrid memory for eval tasks (#21224)

mudler committed 67 days ago

Verified e1cb8174

CUDA/HIP: Fix kernel slection for mmvq mmid kernel to align host selection with device launch bounds (#21238)

IMbackK committed 67 days ago

Verified 88d5f8ff

ggml : fix RWKV ops thread assignment (#21226)

ggerganov committed 67 days ago

Verified d43375ff

ggml-cpu: fix fallback for RVV kernels without zvfh (#21157)

taimur-10x committed 67 days ago

Verified 2b86e5ca

CUDA: Add Flash Attention Support for Head Dimension 512 (#20998)

anavp-nvidia committed 67 days ago

Verified 88458164

llama : refactor llama_model_quantize_params to expose a pure C interface (#20346)

EAddario committed 67 days ago

Verified 49512502

ggml webgpu: quantized buffers to u32 + wider browser/device support (#21046)

reeselevine committed 67 days ago

Verified 82764c34

ggml-webgpu: port all AOT operators to JIT (#20728)

abhijitramesh committed 68 days ago

Verified 825eb91a

fix: Use lower-case proxy headers naming (#21235)

allozaur committed 68 days ago

Verified 0fcb3760

common : cleanup logs and modernize the progress bar (#21215)

angt committed 68 days ago

Verified 6307ec07

CANN: fix multi-thread set_tensor race conditions (#20151)

hipudding committed 68 days ago

Verified 632219af

server: (webui) no more gzip compression (#21073)

ngxson committed 68 days ago

Verified 4a00bbfe

common : gpt-oss handle builtin and unsolicited tool calls (#21213)

aldehir committed 68 days ago

Verified 624733d6

fix: correct misspellings in code comments (#21217)

lainon1 committed 68 days ago

Verified 0b6ff479

CI: Enable CPU and Vulkan ARM64 Release (#21207)

ehfd committed 68 days ago

Verified eec6f85d

ggerganov committed 68 days ago

9281dd13

ggml : bump version to 0.9.9 (ggml/1449)

ggerganov committed 68 days ago

0be6c7c9

common : move up common_init() and fix Windows UTF-8 logs (#21176)

angt committed 68 days ago

Verified 41361c85

sycl : enhance fattn perf (#21185)

arthw committed 68 days ago

Verified 62278ced

common: add bounds check in common_init_result::sampler to prevent segfault on failed model load (#21082)

mtmcp committed 68 days ago

Verified 90aa83c6

fix: include API key in CORS proxy requests for MCP connections (#21193)

satishkc7 committed 68 days ago

Verified fcc2d598

server/webui: cleanup dual representation approach, simplify to openai-compat (#21090)

pwilkin committed 68 days ago

Verified 4453e775

vendor : update BoringSSL to 0.20260327.0 (#21211)

angt committed 68 days ago

Verified 26dac845

common : Disable backend sampling if reasoning budget is enabled (#21209)

Galunid committed 68 days ago

Verified 5ce013cd

opencl: add q4_K gemm and gemv kernels for Adreno (#20919)

shaofeiqi committed 69 days ago

Verified 08f21453

CI : Enable CUDA and Vulkan ARM64 runners and fix CI/CD (#21122)

ehfd committed 69 days ago

Verified 84ae8434

Newer Older