Repository navigation
Eval bug: segfault when loading ternary bonsai2 #224
Description
Name and Version
version: 0.2.0-dev (build 10709, commit 9a9394a)
Operating systems
Linux
GGML backends
Vulkan
Hardware
AMD Strix Halo
Models
Ternary-Bonsai-2-27B-PQ2_0.gguf
Problem description & steps to reproduce
Loading model... |Segmentation fault (core dumped) ./llama-prism-b10709-9a9394a/llama-cli -m Ternary-Bonsai-2-27B-PQ2_0.gguf -ngl 99 -fa on -c 32768 --temp 1.0 --top-p 0.95 --top-k 20 -p "Explain quantum computing in simple terms." -n 256
First Bad Commit
No response
Relevant log output
cmd line: ./llama-prism-b10709-9a9394a/llama-cli -m Ternary-Bonsai-2-27B-PQ2_0.gguf -ngl 99 -fa on -c 32768 --temp 1.0 --top-p 0.95 --top-k 20 -p "Explain quantum computing in simple terms." -n 256
coredump is 5 GB, too large to share ...
Activity
bri-prism commentedon Sep 24, 2026
Thanks for reporting this. It's the same crash as #180, fixed by #245 and included in the new release prism-b10735-842b188. Closing as a duplicate of #180 so we track it in one place. If the new build still crashes for you, please reopen or comment on #180 with the output.