Lê Nam Khánh
c2324b8f3e
docs: fix typos in libs/langchain/langchain_classic/chains/summarize/chain.py ( #33877 )
2025-11-07 10:33:53 -05:00
Lê Nam Khánh
957ea65d12
docs: fix typos in libs/core/tests/unit_tests/indexing/test_hashed_document.py ( #33874 )
2025-11-07 10:32:20 -05:00
Lê Nam Khánh
00fa38a295
docs: fix typos in libs/core/tests/unit_tests/test_tools.py ( #33876 )
2025-11-07 10:31:57 -05:00
Lê Nam Khánh
9d98c1b669
docs: fix typos in libs/partners/groq/langchain_groq/chat_models.py ( #33878 )
2025-11-07 10:31:35 -05:00
Mahmut CAVDAR
00cc9d421f
fix(langchain): Update langchain-core dependency version ( #33775 )
2025-11-07 10:31:06 -05:00
Mohammad Mohtashim
65716cf590
feat(perplexity): Created Dedicated Output Parser to Support Reasoning Model Output for perplexity ( #33670 )
2025-11-07 10:17:35 -05:00
riunyfir
1b77a191f4
feat: The response.incomplete event is not handled when using stream_mode=['messages'] ( #33871 )
2025-11-07 09:46:11 -05:00
Lê Nam Khánh
2fe0369049
docs: fix typos in some files ( #33867 )
2025-11-07 09:04:29 -05:00
Mason Daugherty
e023201d42
style: some cleanup ( #33857 )
2025-11-06 23:50:46 -05:00
Mason Daugherty
d40e340479
chore: attribute package change versions ( #33854 )
...
Needed to disambiguate for within inherited docs
2025-11-06 16:57:30 -05:00
Sydney Runkle
9a09ed0659
fix: don't trace conditional edges and no todos in input state ( #33842 )
...
while experimenting w/ todo middleware
| Before | After |
|--------|-------|
| 
| 
|
| Tracing conditional edges (verbose) | Not tracing conditional edges
(cleaner) |
| 
| 
|
| Todos in input state (cluttered) | No todos in input state (cleaner) |
2025-11-05 14:25:57 -05:00
Mason Daugherty
022fdd52c3
fix(core): handle missing dependency version information ( #33844 )
...
Follow up to #33347
This continues to make searching issues difficult
2025-11-05 14:19:55 -05:00
Sydney Runkle
7946a8f64e
release: langchain v1.0.4 ( #33839 )
2025-11-05 12:37:58 -05:00
Sydney Runkle
7af79039fc
fix: only increment thread count on successful executions ( #33837 )
...
* for run count + thread count overflow we should warn model not to call
again
* don't tally mocked tool calls in thread limit -- consider the
following
* run limit is 1
* thread limit is 3
* first run calls the tool 2 times, 1 executes, 1 is blocked
* we should only count the successful execution above towards the total
thread count
* raise more helpful warnings on invalid config
2025-11-05 10:00:07 -05:00
Sydney Runkle
1755750ca1
fix: more robust tool call limit middleware ( #33817 )
...
* improving typing (covariance)
* adding in support for continuing w/ tool calls not yet at threshold,
switching default to continue
* moving all logic into after model
```py
ExitBehavior = Literal["continue", "error", "end"]
"""How to handle execution when tool call limits are exceeded.
- `"continue"`: Block exceeded tools with error messages, let other tools continue (default)
- `"error"`: Raise a `ToolCallLimitExceededError` exception
- `"end"`: Stop execution immediately, injecting a ToolMessage and an AI message
for the single tool call that exceeded the limit. Raises `NotImplementedError`
if there are multiple tool calls
"""
```
2025-11-05 09:18:21 -05:00
Mason Daugherty
47d89b1e47
fix(langchain): remove Tigris ( #33829 )
...
Removing this code as there is no possible way for it to work.
See https://github.com/langchain-ai/langchain-community/pull/159
2025-11-04 19:45:52 -05:00
Mason Daugherty
ee0bdaeb79
chore: correct langchain-community references ( #33827 )
...
fix docstrings that referenced community versions of now-native packages
2025-11-04 17:01:35 -05:00
Christophe Bornet
915c446c48
chore(core): add ruff rule PLR2004 ( #33706 )
...
Co-authored-by: Mason Daugherty <mason@langchain.dev >
2025-11-04 13:33:37 -05:00
Mason Daugherty
d1e2099408
chore(core): clean pyproject formatting ( #33821 )
2025-11-04 18:21:15 +00:00
Mason Daugherty
6ea15b9efa
docs(model-profiles): fix typo ( #33820 )
2025-11-04 18:19:55 +00:00
Mason Daugherty
c6547f58b7
style(standard-tests): refs pass ( #33814 )
2025-11-04 00:01:16 -05:00
Mason Daugherty
dfb05a7fa0
style: refs pass ( #33813 )
2025-11-03 22:11:10 -05:00
ccurme
2f67f9ddcb
release(huggingface): 1.0.1 ( #33803 )
2025-11-03 14:49:52 -05:00
Hyejeong Jo
0e36185933
fix(huggingface): add stream_usage support for ChatHuggingFace invoke/stream ( #32708 )
2025-11-03 14:44:32 -05:00
Michael Li
6617865440
fix(core): add no colors check ( #33780 )
...
Patch edge case in get_color_mapping
2025-11-03 13:23:23 -05:00
ccurme
6dba4912be
release(model-profiles): 0.0.3 ( #33798 )
2025-11-03 11:17:08 -05:00
ccurme
7a3827471b
fix(model-profiles): fix pdf_inputs field ( #33797 )
2025-11-03 11:10:33 -05:00
ccurme
f006bc4c7e
feat(langchain): add model-profiles as optional dependency ( #33794 )
2025-11-03 10:13:58 -05:00
Mason Daugherty
0a442644e3
test(anthropic): add vcr to test_search_result_tool_message ( #33793 )
...
To fix nondeterministic results causing integration testing to sometimes
fail
Also speeds up from 10s to 0.5
---------
Co-authored-by: ccurme <chester.curme@gmail.com >
2025-11-03 15:13:30 +00:00
ccurme
1381137c37
release(standard-tests): 1.0.1 ( #33792 )
2025-11-03 09:46:39 -05:00
ccurme
b4a042dfc4
release(core): 1.0.3 ( #33768 )
2025-11-03 09:19:32 -05:00
ccurme
81c4f21b52
fix(standard-tests): update multimodal tests ( #33781 )
2025-11-01 16:38:20 -04:00
Mason Daugherty
f2dab562a8
style: misc refs work ( #33771 )
2025-10-31 18:29:53 -04:00
ccurme
61196a8280
release(openai): 1.0.2 ( #33769 )
2025-10-31 14:21:32 -04:00
ccurme
7a97c31ac0
release(model-profiles): 0.0.2 ( #33767 )
2025-10-31 13:58:04 -04:00
ccurme
424214041e
feat(model-profiles): support more providers ( #33766 )
2025-10-31 13:48:56 -04:00
ccurme
b06bd6a913
fix(model-profiles): add typing-extensions as explicit dep ( #33762 )
2025-10-31 11:21:55 -04:00
ccurme
1c762187e8
fix(model-profiles): remove langchain-core as a dependency ( #33761 )
2025-10-31 11:04:14 -04:00
Mason Daugherty
90aefc607f
docs(core): improve tools module docstrings ( #33755 )
...
styling in `base.py`, content updates in
`libs/core/langchain_core/tools/convert.py`
2025-10-31 10:54:30 -04:00
ccurme
493be259c3
feat(core): mint langchain-model-profiles and add profile property to BaseChatModel ( #33728 )
2025-10-31 09:44:46 -04:00
Mason Daugherty
7aaaa371e7
release(anthropic): 1.0.1 ( #33752 )
2025-10-30 16:19:44 -04:00
Mason Daugherty
468dad1780
chore: use model IDs, latest anthropic models ( #33747 )
...
- standardize on using model IDs, no more aliases - makes future
maintenance easier
- use latest models in docstrings to highlight support
- remove remaining sonnet 3-7 usage due to deprecation
Depends on #33751
2025-10-30 16:13:28 -04:00
Mason Daugherty
32d294b89a
fix(anthropic): clean up tests, update default model to use ID ( #33751 )
...
- use latest models in examples to highlight support
- standardize on using IDs in examples - no more aliases to improve
determinism in future tests
- bump lock
- in integration tests, fix stale casettes and use `MODEL_NAME`
uniformly where possible
- add case for default max tokens for sonnet-4-5 (was missing)
2025-10-30 16:08:18 -04:00
Mason Daugherty
dc5b7dace8
test(openai): mark tests flaky ( #33750 )
...
see:
https://github.com/langchain-ai/langchain/actions/runs/18921929210/job/54020065079#step:10:560
2025-10-30 16:07:58 -04:00
Mason Daugherty
e00b7233cf
chore(langchain): fix lint_imports paths ( #33749 )
2025-10-30 16:06:08 -04:00
Mason Daugherty
91f7e73c27
fix(langchain): use system_prompt in integration tests ( #33748 )
2025-10-30 16:05:57 -04:00
Shagun Gupta
75fff151e8
fix(openai): replace pytest.warns(None) with warnings.catch_warnings in ChatOpenAI test to resolve TypeError . Resolves issue #33705 ( #33741 )
2025-10-30 09:22:34 -04:00
Sydney Runkle
fabcacc3e5
chore: remove mentions of sonnet 3.5 ( #33737 )
...
see
https://docs.claude.com/en/docs/about-claude/model-deprecations#2025-08-13%3A-claude-sonnet-3-5-models
2025-10-29 15:49:27 -07:00
Christian Bromann
ac58d75113
fix(langchain_v1): remove thread_model_call_count and run_model_call_count from tool node test ( #33725 )
...
While working on ToolRuntime in TS I discovered that Python still uses
`thread_model_call_count` and `run_model_call_count` in ToolNode tests
which afaik we removed.
2025-10-29 15:36:18 -07:00
Sydney Runkle
28564ef94e
release: core 1.0.2 and langchain 1.0.3 ( #33736 )
2025-10-29 15:30:17 -07:00
