]> git.djapps.eu Git - pkg/ggml/sources/llama.cpp/commit
overview / pkg / ggml / sources / llama.cpp / commit
summary | shortlog | log | commit | commitdiff | tree
(parent: 2e42be4)
llama-chat : Do not throw when tool parsing fails (#14012)
authorPiotr <redacted>
Sat, 14 Jun 2025 16:25:15 +0000 (18:25 +0200)
committerGitHub <redacted>
Sat, 14 Jun 2025 16:25:15 +0000 (17:25 +0100)
commit3cb203c89f60483e349f841684173446ed23c28f
tree3e3906df05b0c571b9f54c9b7d41151fd314c562tree
parent2e42be42bd6bf1dcc643d6ac4e77419bfe5dd24fcommit | diff
llama-chat : Do not throw when tool parsing fails (#14012)

Currently when a model generates output which looks like a tool call,
but is invalid an exception is thrown and not handled, causing the cli
or llama-server to bail. Instead, handle the chat parser exception and
simply return the generated text in such cases.

Signed-off-by: Piotr Stankiewicz <redacted>
common/chat-parser.cpp diff | blob | history
common/chat-parser.h diff | blob | history
common/chat.cpp diff | blob | history
Packaging of ggml-org/llama.cpp