All issues

Search results

Misc. bug: Context checkpoints not working

Status: Open.

#396 In TheTom/llama-cpp-turboquant;

· jankais3r opened on Sep 29, 2026

Permission to submit your turbo KV (HIP port) to halo-box/strix-llama.cpp, credited to you

Status: Open.

#394 In TheTom/llama-cpp-turboquant;

· BonkusCheemus opened on Sep 28, 2026

Eval bug: --kv-stream-arena-mib crashes with illegal memory access on Ada (sm_89) at load time

Status: Open.

#393 In TheTom/llama-cpp-turboquant;

· stuarts0011 opened on Sep 28, 2026

kv-stream: native turbo KV streaming - port offer with bench evidence (decode 2.42 -> 15 t/s at deep spill)

Status: Open.

#391 In TheTom/llama-cpp-turboquant;

· grogroningem-code opened on Sep 26, 2026

CUDA: SET_ROWS_TURBO3/TURBO4 tests never run because CPY from turbo to F32 is unsupported

Status: Open.

#385 In TheTom/llama-cpp-turboquant;

· TheTom opened on Sep 23, 2026

Misc. bug: -40% decode speed regression with DeepSeek V4 Flash

Status: Open.

#379 In TheTom/llama-cpp-turboquant;

· Volunteer-1 opened on Sep 20, 2026

speculative-simple: null draft context with a DFlash2 drafter (dflash requires ctx_other to be set)

Status: Open.

#376 In TheTom/llama-cpp-turboquant;

· jasstrong opened on Sep 20, 2026

Vulkan MoE cache: 512 KiB min-expert default silently excludes small-expert MoE models (Qwen3.6-35B-A3B IQ2_XS, 296 KiB/expert)

Status: Open.

#365 In TheTom/llama-cpp-turboquant;

· Scintsz opened on Sep 9, 2026

SYCL: TurboQuant flash-attn decode (vec) kernels fail JIT on Arc Pro B60 (G21, 0xe211) with head_dim 256 models - pp OK, tg aborts

Status: Open.

#361 In TheTom/llama-cpp-turboquant;

· fantaskiss opened on Sep 7, 2026

Eval bug: First decode after prefill runs 2-45x slower: HIP CUDA-graph capture warmup (Qwen3.8-Flash-Next / qwen4exp QSA, gfx1201)

Status: Open.

#356 In TheTom/llama-cpp-turboquant;

· Draco-Lunaris opened on Sep 5, 2026

CUDA: full test-backend-ops sweep segfaults in ggml_backend_cuda_graph_compute at ROLL on GB10 (passes with graphs disabled)

bugSomething isn't workingSomething isn't working

Status: Open.

#347 In TheTom/llama-cpp-turboquant;

· TheTom opened on Sep 3, 2026

Proposal: TurboAnchorKV cold prompt KV experiment

Status: Open.

#332 In TheTom/llama-cpp-turboquant;

· TheTom opened on Aug 31, 2026