./llama-b9568$ ./llama-server -hf ggml-org/Qwen3-Omni-30B-A3B-Instruct-GGUF:Q4_K_M --threads -1 --parallel 1 --ctx-size 16384 --temp 0.6 --top-p 0.95 --top-k 20 --predict 4096 --direct-io --log-verbosity 9999
Input mp4 file 11 minutes long over the web ui.
0.22.805.922 D === Differential analysis complete ===
0.22.806.329 D common_chat_templates_apply_jinja: generated parser:
Sequence(Literal(<|im_start|>assistant
), Space, Repetition(Sequence(Epsilon, Literal(<think>), Repetition(Literal(
), 0, 1), Tag(reasoning, Until(</think>)), Epsilon, Repetition(Literal(
), 0, 1), Literal(</think>), Repetition(Literal(
), 0, 1), Repetition(Literal(
), 0, 1)), 0, 1), Tag(content, Until()), End)
parser generation prompt: <|im_start|>assistant
<think>
</think>
0.22.829.787 D probe: launching:0.22.829.792 D ffprobe0.22.829.792 D -v0.22.829.793 D quiet0.22.829.793 D -show_entries0.22.829.794 D stream=width,height,r_frame_rate,nb_frames,duration0.22.829.794 D -select_streams0.22.829.794 D v:00.22.829.794 D -of0.22.829.794 D default=noprint_wrappers=10.22.829.795 D pipe:00.22.829.795 D
0.22.832.198 D feed_stdin: feeding 22449655 bytes to stdin
0.22.885.258 D feed_stdin: wrote 22449655 bytes, closing stdin
corrupted double-linked list
Aborted
Name and Version
.
/llama-b9568$ ./llama-server -hf ggml-org/Qwen3-Omni-30B-A3B-Instruct-GGUF:Q4_K_M --threads -1 --parallel 1 --ctx-size 16384 --temp 0.6 --top-p 0.95 --top-k 20 --predict 4096 --direct-io --log-verbosity 9999Input mp4 file 11 minutes long over the web ui.
log output of first error: corrupted double-linked list
Operating systems
Linux
GGML backends
Vulkan
Hardware
ryzen 7840U
Models
No response
Problem description & steps to reproduce
add a mp4 file for transcription
First Bad Commit
No response
Relevant log output
Logs