tqcq air

tqcq synced commits to refs/heads/xsn/common_subproc at ggml-org/llama.cpp from mirror 2026-07-25 13:29:49 +08:00

tqcq synced new reference refs/tags/b10107 to ggml-org/llama.cpp from mirror 2026-07-25 05:19:32 +08:00

tqcq synced commits to refs/tags/b10107 at ggml-org/llama.cpp from mirror 2026-07-25 05:19:32 +08:00

tqcq synced new reference refs/heads/0cc4m/mmap-auto to ggml-org/llama.cpp from mirror 2026-07-25 05:19:32 +08:00

tqcq synced commits to refs/heads/0cc4m/mmap-auto at ggml-org/llama.cpp from mirror 2026-07-25 05:19:32 +08:00

tqcq synced commits to refs/heads/xsn/server_mcp_stdio at ggml-org/llama.cpp from mirror 2026-07-25 05:19:32 +08:00

6dfc0857fc fix some edge cases

a669ac5a55 Merge branch 'master' into xsn/server_mcp_stdio

49d0ad7b98 internal/mcp-stdio: integration + tests + fixes (#26075)

298219f985 llama: various bug fixes (#26051)

fa72aeccb2 HIP: remove rocWMMA FlashAttention (#26046)

Compare 11 commits »

tqcq synced commits to refs/heads/master at ggml-org/llama.cpp from mirror 2026-07-25 05:19:32 +08:00

555881ebc8 ui: reduce per-token render cost when streaming (#26053)

96013c5112 ui: remove render effects (#26083)

88bfee1429 model: add GLM 5.2 Indexer support (#25407)

95a923a64c ui: fix MCP server display name conflicts in tools lists (#26011)

27209a598d server: support "reasoning_effort": "none" in OAI API (#26045)

Compare 11 commits »

tqcq synced commits to refs/tags/b10106 at ggml-org/llama.cpp from mirror 2026-07-24 21:09:39 +08:00

tqcq synced new reference refs/tags/b10105 to ggml-org/llama.cpp from mirror 2026-07-24 21:09:39 +08:00

tqcq synced commits to refs/tags/b10105 at ggml-org/llama.cpp from mirror 2026-07-24 21:09:39 +08:00

tqcq synced new reference refs/tags/b10106 to ggml-org/llama.cpp from mirror 2026-07-24 21:09:39 +08:00

tqcq synced new reference refs/tags/b10108 to ggml-org/llama.cpp from mirror 2026-07-24 21:09:39 +08:00

tqcq synced commits to refs/tags/b10108 at ggml-org/llama.cpp from mirror 2026-07-24 21:09:39 +08:00

tqcq synced new reference refs/heads/xsn/server_mcp_stdio to ggml-org/llama.cpp from mirror 2026-07-24 21:09:39 +08:00

tqcq synced commits to refs/heads/xsn/server_mcp_stdio at ggml-org/llama.cpp from mirror 2026-07-24 21:09:39 +08:00

tqcq synced commits to refs/heads/master at ggml-org/llama.cpp from mirror 2026-07-24 21:09:39 +08:00

8f5ab832ca cohere2 moe template parser: enforce JSON schema for text responses if a response schema is provided (#26018)

0cea36222f vendor: update subprocess.h (#26061)

Compare 2 commits »

tqcq synced new reference refs/heads/xsn/bump_subproc to ggml-org/llama.cpp from mirror 2026-07-24 21:09:39 +08:00

tqcq synced commits to refs/heads/xsn/bump_subproc at ggml-org/llama.cpp from mirror 2026-07-24 21:09:39 +08:00

tqcq synced commits to refs/heads/cross-profiler at ggml-org/llama.cpp from mirror 2026-07-24 21:09:39 +08:00

df6daabc67 Add tensor name to JSON output

9ed6048dd9 tentative Metal support

2ade1b825a Add missing unrolls

695dd1ea54 Revert accidental change.

f1c08eac44 Fix braces

Compare 16 commits »

tqcq synced commits to refs/heads/master at ggml-org/llama.cpp from mirror 2026-07-24 13:00:12 +08:00

0a50d9909a hexagon: further improved pipeline of the core bits (L2, DMA, MM, FA) (#26049)