langchain-ai/deepagents - CodSpeed

deepagents

Blog Docs Changelog

Performance History

Latest Results

test(bench): resize workload to realistic token footprint Previous params (30KB×2 writes + 80KB tool result every turn) accumulated ~10M tokens of state at N=200 — an unrealistic stress test. Resized to ~500k tokens at N=200: 1 KB file write every turn, small (2 KB) tool results most turns, with a larger 82 KB result every 10th turn that still triggers FilesystemMiddleware eviction into the `files` channel so the eviction path stays exercised. Also enables N=200 on async mode now that baseline peak fits (was OOMing at the old payload size). Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>

sr/delta-channel-perf-combo

29 minutes ago

fix(repl,cli): pass tools=[] to ToolRuntime ctors Also relax the multi_turn_eviction xfail to strict=False — the bug only reproduces on macOS; Linux CI passes the test under DeltaChannel. Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>

sr/delta-channel-perf-combo

57 minutes ago

test: xfail multi_turn_eviction pending upstream DeltaChannel fix DeltaChannel(add_messages) replay re-applies each step delta through add_messages, but across-step ID dedup isn't preserved — so an evicted HumanMessage and its replacement both survive reconstruction. The test asserts the replacement alone should win, which is correct behaviour against the baseline reducer. Mark xfail strict=True so we notice once the upstream fix lands. Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>

sr/delta-channel-perf-combo

1 hour ago

fix(tests,middleware): pass tools=[] to ToolRuntime ctors; use get_state API The ToolNode hydrate-from-channels change added a required `tools` arg to `ToolRuntime.__init__`; update all test fixtures and the four production middlewares that build a synthetic ToolRuntime (summarization / filesystem / memory / skills) to pass `tools=[]`. Memory + skills default-backend tests now use `agent.get_state(config).values` instead of reading `checkpoint['channel_values'][...]` directly — DeltaChannel stores a sentinel in the raw checkpoint, and get_state resolves it. Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>

sr/delta-channel-perf-combo

1 hour ago

chore: track DeltaChannel rename to private _delta module Upstream langgraph moved channels/delta.py → channels/_delta.py and removed the public re-export to signal the API is still experimental. Update the FilesystemState import and refresh pinned commits on both combo branches. Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>

sr/delta-channel-perf-combo

1 hour ago

chore: drop local planning artifacts (design lives in docs PR)

vic/deepagent-deploy-frontend-v2

6 hours ago

revert: drop AGENTS.md + linked-entry write gates — defer ACL to higher layer

vishnu/hub-backend

8 hours ago

docs: tighten _IMMUTABLE_BASENAMES comment

vishnu/hub-backend

9 hours ago

Latest Branches

0%

perf(sdk): `DeltaChannel` + `add_messages` fast-path + no-inline Sends#2910

32 minutes ago

2577e30

sr/delta-channel-perf-combo

0%

feat(cli): bundled chat frontend for deepagent deploy#2924

6 hours ago

0f9c52c

vic/deepagent-deploy-frontend-v2

-1%

feat(sdk): add `ContextHubBackend` for LangSmith Hub agent repos#2898

8 hours ago

aa81711

vishnu/hub-backend

© 2026 CodSpeed Technology

Home Terms Privacy Docs