BerriAI
litellm
Blog
Docs
Changelog
Blog
Docs
Changelog
Overview
Branches
Benchmarks
Runs
perf(streaming): cut per-chunk overhead ~30% on Anthropic + Bedrock hot path
#28720
Merged
Comparing
litellm_fix/LIT-3313-streaming-chunk-overhead
(
e8bf753
) with
litellm_internal_staging
(
35f6961
)
CodSpeed Performance Gauge
0%
Untouched
16
Benchmarks
Mode
CPU Simulation
Wall Time
Memory
Status
Untouched
16 total
Uses the
CPU Simulation instrument
to collect CPU performance metrics.
test_get_model_info_with_provider
tests/benchmarks/test_benchmarks.py
CodSpeed Performance Gauge
-1%
79.8 µs
81 µs
Uses the
CPU Simulation instrument
to collect CPU performance metrics.
test_get_model_info_anthropic
tests/benchmarks/test_benchmarks.py
CodSpeed Performance Gauge
+1%
80.9 µs
80 µs
Uses the
CPU Simulation instrument
to collect CPU performance metrics.
test_get_llm_provider_openai
tests/benchmarks/test_benchmarks.py
CodSpeed Performance Gauge
-1%
138 µs
139.4 µs
Uses the
CPU Simulation instrument
to collect CPU performance metrics.
test_token_counter_long_content
tests/benchmarks/test_benchmarks.py
CodSpeed Performance Gauge
-1%
1.7 ms
1.7 ms
Uses the
CPU Simulation instrument
to collect CPU performance metrics.
test_token_counter_simple_message
tests/benchmarks/test_benchmarks.py
CodSpeed Performance Gauge
+1%
236.7 µs
235.1 µs
Uses the
CPU Simulation instrument
to collect CPU performance metrics.
test_cost_per_token_openai
tests/benchmarks/test_benchmarks.py
CodSpeed Performance Gauge
0%
554.8 µs
556.7 µs
Uses the
CPU Simulation instrument
to collect CPU performance metrics.
test_get_model_cost_key_case_insensitive
tests/benchmarks/test_benchmarks.py
CodSpeed Performance Gauge
0%
82.2 µs
82 µs
Uses the
CPU Simulation instrument
to collect CPU performance metrics.
test_token_counter_with_tools
tests/benchmarks/test_benchmarks.py
CodSpeed Performance Gauge
0%
413.7 µs
412.6 µs
Uses the
CPU Simulation instrument
to collect CPU performance metrics.
test_token_counter_raw_text
tests/benchmarks/test_benchmarks.py
CodSpeed Performance Gauge
0%
183.9 µs
183.5 µs
Uses the
CPU Simulation instrument
to collect CPU performance metrics.
test_cost_per_token_anthropic
tests/benchmarks/test_benchmarks.py
CodSpeed Performance Gauge
0%
556.5 µs
557.8 µs
Uses the
CPU Simulation instrument
to collect CPU performance metrics.
test_token_counter_multi_turn
tests/benchmarks/test_benchmarks.py
CodSpeed Performance Gauge
0%
566.9 µs
567.7 µs
Uses the
CPU Simulation instrument
to collect CPU performance metrics.
test_get_model_info_openai
tests/benchmarks/test_benchmarks.py
CodSpeed Performance Gauge
0%
81.9 µs
81.7 µs
Uses the
CPU Simulation instrument
to collect CPU performance metrics.
test_get_llm_provider_with_prefix
tests/benchmarks/test_benchmarks.py
CodSpeed Performance Gauge
0%
136.9 µs
137 µs
Uses the
CPU Simulation instrument
to collect CPU performance metrics.
test_get_llm_provider_anthropic
tests/benchmarks/test_benchmarks.py
CodSpeed Performance Gauge
0%
142.5 µs
142.5 µs
Uses the
CPU Simulation instrument
to collect CPU performance metrics.
test_get_model_cost_key_exact_match
tests/benchmarks/test_benchmarks.py
CodSpeed Performance Gauge
0%
79.9 µs
79.9 µs
Uses the
CPU Simulation instrument
to collect CPU performance metrics.
test_get_llm_provider_azure
tests/benchmarks/test_benchmarks.py
CodSpeed Performance Gauge
0%
144.2 µs
144.2 µs
Commits
Click on a commit to change the comparison range
Base
main
35f6961
+56.72%
chore: update uv.lock after dependency resolution
42edc8c
25 days ago
by claude
-56.81%
fix(ci): restore uv.lock generated by uv 0.10.9 for lint compatibility
e8bf753
25 days ago
by claude
© 2026 CodSpeed Technology
Home
Terms
Privacy
Docs