CS348Project/llama.cpp
5
0
Fork 0
mirror of https://github.com/ggml-org/llama.cpp.git synced 2025-10-28 08:31:25 +00:00
Code Issues Packages Projects Releases Wiki Activity

server : fix --jinja when there's no tools or schema (typo was forcing JSON) (#11531)

Browse Source
This commit is contained in:
Olivier Chafik
2025-01-31 08:12:40 +00:00
committed by GitHub
parent 1bd3047a93
commit 4a2b196d03
2 changed files with 4 additions and 5 deletions
Download Patch File Download Diff File Expand all files Collapse all files

7
examples/server/tests/unit/test_chat_completion.py
View File

@@ -14,11 +14,10 @@ def create_server():
"model,system_prompt,user_prompt,max_tokens,re_content,n_prompt,n_predicted,finish_reason,jinja,chat_template",
[
(None, "Book", "What is the best book", 8, "(Suddenly)+|\\{ \" Sarax.", 77, 8, "length", False, None),
(None, "Book", "What is the best book", 8, "(Suddenly)+|\\{ \" Sarax.", 77, 8, "length", True, None),
(None, "Book", "What is the best book", 8, "^ blue", 23, 8, "length", True, "This is not a chat template, it is"),
("codellama70b", "You are a coding assistant.", "Write the fibonacci function in c++.", 128, "(Aside|she|felter|alonger)+", 104, 64, "length", False, None),
# TODO: fix testing of non-tool jinja mode
# (None, "Book", "What is the best book", 8, "(Suddenly)+|\\{ \" Sarax.", 77, 8, "length", True, None),
# (None, "Book", "What is the best book", 8, "I want to play with", 23, 8, "length", True, "This is not a chat template, it is"),
# ("codellama70b", "You are a coding assistant.", "Write the fibonacci function in c++.", 128, "(Aside|she|felter|alonger)+", 104, 64, "length", True, None),
("codellama70b", "You are a coding assistant.", "Write the fibonacci function in c++.", 128, "(Aside|she|felter|alonger)+", 104, 64, "length", True, None),
]
)
def test_chat_completion(model, system_prompt, user_prompt, max_tokens, re_content, n_prompt, n_predicted, finish_reason, jinja, chat_template):

2
examples/server/utils.hpp
View File

@@ -642,7 +642,7 @@ static json oaicompat_completion_params_parse(
inputs.parallel_tool_calls = json_value(body, "parallel_tool_calls", false);
inputs.stream = stream;
// TODO: support mixing schema w/ tools beyond generic format.
inputs.json_schema = json_value(llama_params, "json_schema", json::object());
inputs.json_schema = json_value(llama_params, "json_schema", json());
auto chat_params = common_chat_params_init(tmpl, inputs);
llama_params["chat_format"] = static_cast<int>(chat_params.format);