fix: [Bug]: Health checks use max_completion_tokens=1, causing failures for GPT-5 models (#23836)
#24893ClosedComparing
hannahmadison:git-genie/issue-23836
(c62a1e2
) withlitellm_oss_branch
(09cd7e3
)0%
- Untouched
- 16
Benchmarks
16 total
test_get_model_info_with_provider
tests/benchmarks/test_benchmarks.py+1%
81.2 µs80.4 µs
test_token_counter_long_content
tests/benchmarks/test_benchmarks.py+1%
1.7 ms1.7 ms
test_token_counter_raw_text
tests/benchmarks/test_benchmarks.py-1%
182.7 µs184.3 µs
test_get_llm_provider_openai
tests/benchmarks/test_benchmarks.py-1%
138.2 µs138.9 µs
test_token_counter_with_tools
tests/benchmarks/test_benchmarks.py0%
411.4 µs413.2 µs
test_get_model_cost_key_exact_match
tests/benchmarks/test_benchmarks.py0%
79.7 µs80 µs
test_get_model_info_openai
tests/benchmarks/test_benchmarks.py0%
80.7 µs80.5 µs
test_get_model_info_anthropic
tests/benchmarks/test_benchmarks.py0%
79.5 µs79.2 µs
test_cost_per_token_anthropic
tests/benchmarks/test_benchmarks.py0%
543 µs541.7 µs
test_get_model_cost_key_case_insensitive
tests/benchmarks/test_benchmarks.py0%
81.9 µs82.1 µs
test_get_llm_provider_with_prefix
tests/benchmarks/test_benchmarks.py0%
136.8 µs137.1 µs
test_get_llm_provider_azure
tests/benchmarks/test_benchmarks.py0%
144.2 µs143.9 µs
test_token_counter_multi_turn
tests/benchmarks/test_benchmarks.py0%
564.6 µs564 µs
test_get_llm_provider_anthropic
tests/benchmarks/test_benchmarks.py0%
142.2 µs142.1 µs
test_cost_per_token_openai
tests/benchmarks/test_benchmarks.py0%
541.1 µs540.9 µs
test_token_counter_simple_message
tests/benchmarks/test_benchmarks.py0%
235.2 µs235.2 µs
Commits
Click on a commit to change the comparison rangeBase
main
09cd7e3
-0.02%
Fix #23836: [Bug]: Health checks use max_completion_tokens=1, causing failures for GPT-5 models
043bc8e
5 months ago
by hannahmadison+0.05%
Merge branch 'main' into git-genie/issue-23836
c62a1e2
4 months ago
by hannahmadison