b10750
Pre-release
kv-cells: look up the n-gram history in the sequence position index (#28040)
get_prev_tokens() rebuilt a (seq, pos) -> token hash map on every
ubatch by walking all used cells, while llama_kv_cells already keeps
an ordered index of the positions of each sequence in seq_pos, updated
on every cell mutation to serve seq_pos_min() and seq_pos_max().
The index now stores (pos, cell) pairs in a std::set instead of a
position -> count map, so a repeated position (cache reuse via rm + add,
vision inputs with shared positions) yields distinct entries and the
removal of a cell erases its own pair. The new seq_pos_tok_le() returns
the token of the cell at the largest position <= p in logarithmic time,
which is exactly what the old window lookup and its M-RoPE gap fallback
computed together.
get_prev_tokens() shrinks to a direct lookup per (token, offset) and
for_each_token_in() goes away with its only caller. The kv-cache keeps
no n-gram logic of its own.
Measured on Qwen3.8-Flash-Next UD-Q4_K_XL at 71k context, alternating
two binaries with the first run discarded: tg 69.3 -> 72.7 t/s (+4.9%),
pp unchanged at ~2720 t/s, greedy output identical, needle retrieved.
Website:
Attestations:
macOS/iOS:
- macOS Apple Silicon (arm64)
- macOS Apple Silicon (arm64, KleidiAI enabled) DISABLED
- macOS Intel (x64)
- iOS XCFramework
Linux:
- Ubuntu x64 (CPU)
- Ubuntu arm64 (CPU)
- Ubuntu s390x (CPU)
- Ubuntu x64 (Vulkan)
- Ubuntu arm64 (Vulkan)
- Ubuntu x64 (ROCm 7.14)
- Ubuntu x64 (OpenVINO)
- Ubuntu x64 (SYCL FP32)
- Ubuntu x64 (SYCL FP16)
Android:
Windows:
- Windows x64 (CPU)
- Windows arm64 (CPU)
- Windows arm64 (OpenCL Adreno)
- Windows x64 (CUDA 12) - CUDA 12.4 DLLs
- Windows x64 (CUDA 13) - CUDA 13.3 DLLs
- Windows arm64 (CUDA 13) (preview) - CUDA 13.4 DLLs
- Windows x64 (Vulkan)
- Windows x64 (OpenVINO)
- Windows x64 (SYCL)
- Windows x64 (ROCm 7.14)
openEuler:
- DISABLED
- openEuler x86 (310p)
- openEuler x86 (910b, ACL Graph)
- openEuler aarch64 (310p)
- openEuler aarch64 (910b, ACL Graph)
UI: