mirror of
https://github.com/crewAIInc/crewAI.git
synced 2026-05-03 00:02:36 +00:00
fix: prioritize task output_json over LLM response_format
This commit fixes issue #3639 by ensuring task-level output settings (output_json and output_pydantic) take precedence over agent-level LLM response_format when both are set with Pydantic models. Changes: - Modified LLM._prepare_completion_params() to accept from_task parameter and check if task has output_json or output_pydantic set - If task has output settings, LLM's response_format is ignored - Updated LLM.call() to pass from_task to _prepare_completion_params() - Added comprehensive test to verify the priority behavior The fix ensures predictable behavior following the standard configuration hierarchy where more specific (task-level) settings override general (agent-level) defaults. Co-Authored-By: João <joao@crewai.com>
This commit is contained in:
Devin AI
@@ -0,0 +1,131 @@
|
||||
interactions:
|
||||
- request:
|
||||
body: '{"messages": [{"role": "system", "content": "You are Test Agent. You''re an
|
||||
expert in testing priority systems.\nYour personal goal is: Test goal for priority testing\nTo give my best complete final answer to the task use the exact following
|
||||
format:\n\nThought: I now can give a great answer\nFinal Answer: Your final
|
||||
answer must be the great and the most complete as possible, it must be outcome
|
||||
described.\n\nI MUST use these formats, my job depends on it!"}, {"role": "user",
|
||||
"content": "\nCurrent Task: Analyze the priority system and provide a result\n\nThis is the expect criteria
|
||||
for your final answer: A structured result with task_result and task_confidence fields.\nyou MUST return the actual complete
|
||||
content as the final answer, not a summary.\n\nBegin! This is VERY important
|
||||
to you, use the tools available and give your best Final Answer, your job depends
|
||||
on it!\n\nThought:"}], "model": "gpt-4o-mini"}'
|
||||
headers:
|
||||
accept:
|
||||
- application/json
|
||||
accept-encoding:
|
||||
- gzip, deflate
|
||||
connection:
|
||||
- keep-alive
|
||||
content-type:
|
||||
- application/json
|
||||
host:
|
||||
- api.openai.com
|
||||
user-agent:
|
||||
- OpenAI/Python 1.47.0
|
||||
method: POST
|
||||
uri: https://api.openai.com/v1/chat/completions
|
||||
response:
|
||||
content: "{\n \"id\": \"chatcmpl-test123\",\n \"object\":
|
||||
\"chat.completion\",\n \"created\": 1727214471,\n \"model\": \"gpt-4o-mini\",\n
|
||||
\ \"choices\": [\n {\n \"index\": 0,\n \"message\": {\n \"role\":
|
||||
\"assistant\",\n \"content\": \"I now can give a great answer\\nFinal
|
||||
Answer: The priority system is working correctly with task-level settings taking precedence over agent-level settings.\",\n \"refusal\": null\n },\n \"logprobs\": null,\n
|
||||
\ \"finish_reason\": \"stop\"\n }\n ],\n \"usage\": {\n \"prompt_tokens\":
|
||||
186,\n \"completion_tokens\": 25,\n \"total_tokens\": 211,\n \"completion_tokens_details\":
|
||||
{\n \"reasoning_tokens\": 0\n }\n },\n \"system_fingerprint\": \"fp_test\"\n}\n"
|
||||
headers:
|
||||
CF-Cache-Status:
|
||||
- DYNAMIC
|
||||
CF-RAY:
|
||||
- 8c85f9af4ef31cf3-GRU
|
||||
Connection:
|
||||
- keep-alive
|
||||
Content-Type:
|
||||
- application/json
|
||||
Date:
|
||||
- Tue, 24 Sep 2024 21:47:52 GMT
|
||||
Server:
|
||||
- cloudflare
|
||||
openai-organization:
|
||||
- crewai-iuxna1
|
||||
openai-processing-ms:
|
||||
- '170'
|
||||
openai-version:
|
||||
- '2020-10-01'
|
||||
x-ratelimit-limit-requests:
|
||||
- '10000'
|
||||
x-ratelimit-limit-tokens:
|
||||
- '30000000'
|
||||
x-ratelimit-remaining-requests:
|
||||
- '9999'
|
||||
x-ratelimit-remaining-tokens:
|
||||
- '29999781'
|
||||
http_version: HTTP/1.1
|
||||
status_code: 200
|
||||
- request:
|
||||
body: '{"messages": [{"role": "user", "content": "The priority system is working correctly with task-level settings taking precedence over agent-level settings."}, {"role": "system", "content":
|
||||
"I''m gonna convert this raw text into valid JSON.\n\nThe json should have the
|
||||
following structure, with the following keys:\n{\n task_result: str\n task_confidence: float\n}"}], "model":
|
||||
"gpt-4o-mini", "tool_choice": {"type": "function", "function": {"name": "TaskOutputModel"}},
|
||||
"tools": [{"type": "function", "function": {"name": "TaskOutputModel", "description":
|
||||
"Correctly extracted `TaskOutputModel` with all the required parameters with correct
|
||||
types", "parameters": {"properties": {"task_result": {"title": "Task Result", "type": "string"}, "task_confidence": {"title": "Task Confidence", "type": "number"}},
|
||||
"required": ["task_result", "task_confidence"], "type": "object"}}}]}'
|
||||
headers:
|
||||
accept:
|
||||
- application/json
|
||||
accept-encoding:
|
||||
- gzip, deflate
|
||||
connection:
|
||||
- keep-alive
|
||||
content-type:
|
||||
- application/json
|
||||
host:
|
||||
- api.openai.com
|
||||
user-agent:
|
||||
- OpenAI/Python 1.47.0
|
||||
method: POST
|
||||
uri: https://api.openai.com/v1/chat/completions
|
||||
response:
|
||||
content: "{\n \"id\": \"chatcmpl-test456\",\n \"object\":
|
||||
\"chat.completion\",\n \"created\": 1727214472,\n \"model\": \"gpt-4o-mini\",\n
|
||||
\ \"choices\": [\n {\n \"index\": 0,\n \"message\": {\n \"role\":
|
||||
\"assistant\",\n \"content\": null,\n \"tool_calls\": [\n {\n
|
||||
\ \"id\": \"call_test789\",\n \"type\":
|
||||
\"function\",\n \"function\": {\n \"name\": \"TaskOutputModel\",\n
|
||||
\ \"arguments\": \"{\\\"task_result\\\":\\\"The priority system is working correctly with task-level settings taking precedence over agent-level settings.\\\",\\\"task_confidence\\\":0.95}\"\n }\n }\n
|
||||
\ ],\n \"refusal\": null\n },\n \"logprobs\": null,\n
|
||||
\ \"finish_reason\": \"stop\"\n }\n ],\n \"usage\": {\n \"prompt_tokens\":
|
||||
120,\n \"completion_tokens\": 30,\n \"total_tokens\": 150,\n \"completion_tokens_details\":
|
||||
{\n \"reasoning_tokens\": 0\n }\n },\n \"system_fingerprint\": \"fp_test\"\n}\n"
|
||||
headers:
|
||||
CF-Cache-Status:
|
||||
- DYNAMIC
|
||||
CF-RAY:
|
||||
- 8c85f9b2fc671cf3-GRU
|
||||
Connection:
|
||||
- keep-alive
|
||||
Content-Type:
|
||||
- application/json
|
||||
Date:
|
||||
- Tue, 24 Sep 2024 21:47:52 GMT
|
||||
Server:
|
||||
- cloudflare
|
||||
openai-organization:
|
||||
- crewai-iuxna1
|
||||
openai-processing-ms:
|
||||
- '163'
|
||||
openai-version:
|
||||
- '2020-10-01'
|
||||
x-ratelimit-limit-requests:
|
||||
- '10000'
|
||||
x-ratelimit-limit-tokens:
|
||||
- '30000000'
|
||||
x-ratelimit-remaining-requests:
|
||||
- '9999'
|
||||
x-ratelimit-remaining-tokens:
|
||||
- '29999947'
|
||||
http_version: HTTP/1.1
|
||||
status_code: 200
|
||||
version: 1
|
||||
Reference in New Issue
Block a user