Avatar for the langchain-ai user
langchain-ai
deepagents
BlogDocsChangelog

perf(sdk): `DeltaChannel` + `add_messages` fast-path + no-inline Sends

#2910Closed
Comparing
sr/delta-channel-perf-combo
(
268b855
) with
main
(
f4a2309
)
CodSpeed Performance Gauge
0%
Untouched
32
Skipped
15

Benchmarks

15 total
test_scaling_subagents[10_subagents]
libs/deepagents/tests/benchmarks/test_benchmark_create_deep_agent.py::TestCreateDeepAgentScaling
Skipped
317.4 ms*
test_with_tools
libs/deepagents/tests/benchmarks/test_benchmark_create_deep_agent.py::TestCreateDeepAgentBenchmark
Skipped
54.2 ms*
test_scaling_tools[20_tools]
libs/deepagents/tests/benchmarks/test_benchmark_create_deep_agent.py::TestCreateDeepAgentScaling
Skipped
56.9 ms*
test_scaling_subagents[3_subagents]
libs/deepagents/tests/benchmarks/test_benchmark_create_deep_agent.py::TestCreateDeepAgentScaling
Skipped
132.4 ms*
test_scaling_subagents[5_subagents]
libs/deepagents/tests/benchmarks/test_benchmark_create_deep_agent.py::TestCreateDeepAgentScaling
Skipped
185.3 ms*
test_with_multiple_subagents
libs/deepagents/tests/benchmarks/test_benchmark_create_deep_agent.py::TestCreateDeepAgentBenchmark
Skipped
185.4 ms*
test_filesystem_init
libs/deepagents/tests/benchmarks/test_benchmark_create_deep_agent.py::TestCreateDeepAgentBenchmark
Skipped
359.3 µs*
test_with_one_subagent
libs/deepagents/tests/benchmarks/test_benchmark_create_deep_agent.py::TestCreateDeepAgentBenchmark
Skipped
80.1 ms*
test_create_deep_agent_minimal
libs/deepagents/tests/benchmarks/test_benchmark_create_deep_agent.py::TestCreateDeepAgentBenchmark
Skipped
53.4 ms*
test_scaling_tools[10_tools]
libs/deepagents/tests/benchmarks/test_benchmark_create_deep_agent.py::TestCreateDeepAgentScaling
Skipped
55.4 ms*
test_scaling_tools[5_tools]
libs/deepagents/tests/benchmarks/test_benchmark_create_deep_agent.py::TestCreateDeepAgentScaling
Skipped
54.5 ms*
test_scaling_tools[1_tools]
libs/deepagents/tests/benchmarks/test_benchmark_create_deep_agent.py::TestCreateDeepAgentScaling
Skipped
53.7 ms*
test_scaling_subagents[1_subagents]
libs/deepagents/tests/benchmarks/test_benchmark_create_deep_agent.py::TestCreateDeepAgentScaling
Skipped
80.1 ms*
test_with_string_model_resolution
libs/deepagents/tests/benchmarks/test_benchmark_create_deep_agent.py::TestCreateDeepAgentBenchmark
Skipped
53.7 ms*

Commits

Click on a commit to change the comparison range
Base
main
f4a2309
+10.08%
perf: bundle all 3 perf wins — DeltaChannel + add_messages + no-inline Sends
c852789
1 month ago
by sydney-runkle
-10.76%
test: add bench_perf_combo.py for reproducing perf numbers
95b3c1a
1 month ago
by sydney-runkle
+0.71%
chore: track DeltaChannel rename to private _delta module
9a7e71a
1 month ago
by sydney-runkle
+0.47%
fix(tests,middleware): pass tools=[] to ToolRuntime ctors; use get_state API
e5817db
1 month ago
by sydney-runkle
-0.27%
test: xfail multi_turn_eviction pending upstream DeltaChannel fix
2f9e933
1 month ago
by sydney-runkle
-0.31%
fix(repl,cli): pass tools=[] to ToolRuntime ctors
b4eec91
1 month ago
by sydney-runkle
+0.37%
test(bench): resize workload to realistic token footprint
2577e30
1 month ago
by sydney-runkle
-0.54%
perf(summarization): reuse token count between _truncate_args and _should_summarize
268b855
1 month ago
by sydney-runkle
© 2026 CodSpeed Technology
Home Terms Privacy Docs