Skip to content

Actions: Nexesenex/ik_llama.cpp.nxs

Actions

Python Type-Check

Actions

Loading...
Loading

Show workflow options

Create status badge

Loading
220 workflow runs
220 workflow runs

Filter by Event

⌥ + click/return to exclude

Filter by Status

⌥ + click/return to exclude

Filter by Branch

⌥ + click/return to exclude

Filter by Actor

⌥ + click/return to exclude
ggml-backend: deduplicate output_ready_events waits by source backend… Python Type-Check #237: Commit 9ddf974 pushed by Nexesenex
16s custom_bug_dev
custom_bug_dev
16s
CLI: add --ignore-unknown flag to silently skip unrecognized arguments Python Type-Check #236: Commit 537643e pushed by Nexesenex
15s custom_bug
custom_bug
15s
topk: skip bitmap init/access when K=1 (argmax fast path) Python Type-Check #235: Commit fffbfd0 pushed by Nexesenex
15s custom_bug
custom_bug
15s
quantize: add --protosimd flag for AVX2-accelerated K-quant quantization Python Type-Check #234: Commit 8837106 pushed by Nexesenex
22s custom_bug
custom_bug
22s
Update speculative.cpp Python Type-Check #233: Commit 2e344a9 pushed by Nexesenex
24s custom_bug
custom_bug
24s
Fix -igrm: ban_state s_l/KV cache mismatch causing infinite loop in h… Python Type-Check #232: Commit 0e4069b pushed by Nexesenex
20s custom_bug
custom_bug
20s
Add 2ms sleep after each micro-batch when n_batch > 64 to prevent han… Python Type-Check #231: Commit 24b53a2 pushed by Nexesenex
15s custom_bug
custom_bug
15s
CUDA: apply LCPP PR 15715 norm optimizations to _nc variants and basi… Python Type-Check #230: Commit 9911699 pushed by Nexesenex
18s custom_bug
custom_bug
18s
mmq: per-4-threads restructuring for IQ5_NL Python Type-Check #229: Commit 8573dcb pushed by Nexesenex
22s custom_bug
custom_bug
22s
GGML-CPU: Perfs - O(1) active-expert lookup in chunk-pooling paths Python Type-Check #228: Commit 24245b7 pushed by Nexesenex
17s custom_bug
custom_bug
17s
ggml : fix nb00 undeclared error in im2col_f16 assertion Python Type-Check #227: Commit a6a3fef pushed by Nexesenex
23s custom_neo
custom_neo
23s
Update llama-load-tensors.cpp Python Type-Check #226: Commit c4948d8 pushed by Nexesenex
17s custom_neo
custom_neo
17s
binbcast: fix Q4_0 - use float for d, guard min/max macros Python Type-Check #225: Commit 27fdcbc pushed by Nexesenex
22s custom_neo
custom_neo
22s
Cast reduce result back to critical type after FFN/MoE reduce in Q8_0 Python Type-Check #224: Commit 6415203 pushed by Nexesenex
17s custom_neo
custom_neo
17s
IQ5_NL : IQK implementation. Python Type-Check #223: Commit 8d8f05e pushed by Nexesenex
15s custom_neo
custom_neo
15s
NXS CUDA graphs implementation v1.42 Python Type-Check #222: Commit 05577dc pushed by Nexesenex
21s custom_neo
custom_neo
21s
ggml-cuda: remove unused num_nodes, nodes, params from ggml_cuda_graph Python Type-Check #221: Commit 6f96d9c pushed by Nexesenex
17s custom_neo
custom_neo
17s
ggml-cuda: add LRU eviction for stale CUDA graph instances Python Type-Check #220: Commit d9d8503 pushed by Nexesenex
15s custom_neo
custom_neo
15s
fix: CONV_TRANSPOSE_2D graph_plan work size uses ggml_type_size inste… Python Type-Check #219: Commit 37e34d4 pushed by Nexesenex
17s custom_neo
custom_neo
17s
Q5_0: Add OLS scale refinement to quantize_row_q5_0_ref Python Type-Check #218: Commit e4c7ec9 pushed by Nexesenex
16s custom_neo
custom_neo
16s
fix: sync is_repacked check with k_map in split_buffer get_tensor Python Type-Check #217: Commit 5f4fb9f pushed by Nexesenex
18s custom_neo
custom_neo
18s
Revert "q8_1: full KV cache implementation" Python Type-Check #216: Commit 020ca4d pushed by Nexesenex
22s custom_neo
custom_neo
22s
CUDA: apply LCPP PR 15715 norm optimizations to _nc variants and basi… Python Type-Check #215: Commit c1d85e7 pushed by Nexesenex
23s custom_neo
custom_neo
23s
Update ggml.c Python Type-Check #214: Commit 7a83745 pushed by Nexesenex
23s custom_neo
custom_neo
23s
Refactor: simplify mem_buffer_owned initialization in ggml_init (ggml… Python Type-Check #213: Commit 96034b7 pushed by Nexesenex
22s custom_neo
custom_neo
22s