jdelony

jdelony synced new reference refs/tags/b9785 to jdelony/llama.cpp from mirror 2026-06-24 23:30:21 -05:00

jdelony synced commits to refs/tags/b9786 at jdelony/llama.cpp from mirror 2026-06-24 23:30:21 -05:00

jdelony synced new reference refs/tags/b9786 to jdelony/llama.cpp from mirror 2026-06-24 23:30:21 -05:00

jdelony synced commits to 0cc4m/vulkan-cpp-split at jdelony/llama.cpp from mirror 2026-06-24 23:30:20 -05:00

jdelony synced new reference 0cc4m/vulkan-cpp-split to jdelony/llama.cpp from mirror 2026-06-24 23:30:20 -05:00

jdelony synced commits to caps-harden at jdelony/llama.cpp from mirror 2026-06-24 23:30:20 -05:00

jdelony synced new reference caps-harden to jdelony/llama.cpp from mirror 2026-06-24 23:30:20 -05:00

jdelony synced commits to cisc/quantize-moe-mtp-fix at jdelony/llama.cpp from mirror 2026-06-24 23:30:20 -05:00

jdelony synced new reference cisc/quantize-moe-mtp-fix to jdelony/llama.cpp from mirror 2026-06-24 23:30:20 -05:00

jdelony synced commits to dev-metal at jdelony/llama.cpp from mirror 2026-06-24 23:30:20 -05:00

bec3083830 metal : per-op source split + parallel compile (#24021)

f8cc15f163 [SYCL] support bf16 on bin_bcast OP and unary OPs (#24838)

37957e8531 sampling : remove unconditional softmax+sort in top-n-sigma sampler (#22645)

d0f9d2e5ac server: fix edit_file crash on append at end of file (line_start -1) (#24893)

0ef6f06d55 docs/android.md: Add dependency libandroid-spawn for building in termux (#21812)

Compare 38 commits »

jdelony synced commits to master at jdelony/llama.cpp from mirror 2026-06-24 23:30:20 -05:00

fdb2c11c70 opencl: support non-contig rows in norm (#24965)

09cedfd699 chat: harden caps check (#24973)

8be759e6f7 hexagon: MUL_MAT and MUL_MAT_ID rework : 32x32 tiled weight repack, kernel-params, cached graphs (#24954)

894bb27af3 mtmd: model: unlimited-ocr: converter + parity test (#24969)

fb401045cc common: remove unused json-partial (#24968)

Compare 31 commits »

jdelony synced commits to xsn/cli_http_based at jdelony/llama.cpp from mirror 2026-06-24 23:30:20 -05:00

jdelony synced new reference xsn/cli_http_based to jdelony/llama.cpp from mirror 2026-06-24 23:30:20 -05:00

jdelony synced commits to xsn/rm_unused_json_partial at jdelony/llama.cpp from mirror 2026-06-24 23:30:20 -05:00

jdelony synced new reference xsn/rm_unused_json_partial to jdelony/llama.cpp from mirror 2026-06-24 23:30:20 -05:00

jdelony synced commits to xsn/server_403_disabled_endpoints at jdelony/llama.cpp from mirror 2026-06-24 23:30:20 -05:00

jdelony synced new reference xsn/server_403_disabled_endpoints to jdelony/llama.cpp from mirror 2026-06-24 23:30:20 -05:00

jdelony synced commits to xsn/server_input_file_schema at jdelony/llama.cpp from mirror 2026-06-24 23:30:20 -05:00

1b82e9ae51 fix windows

61653c7989 Merge branch 'master' into xsn/server_input_file_schema

099b579acb ui: model status and load progress via /models/sse feed (#24878)

f8cc15f163 [SYCL] support bf16 on bin_bcast OP and unary OPs (#24838)

37957e8531 sampling : remove unconditional softmax+sort in top-n-sigma sampler (#22645)

Compare 199 commits »

jdelony synced commits to xsn/server_multithread_sampling at jdelony/llama.cpp from mirror 2026-06-24 23:30:20 -05:00

jdelony synced new reference xsn/server_multithread_sampling to jdelony/llama.cpp from mirror 2026-06-24 23:30:20 -05:00