langsmith

What's Changed

• Support parent=false to prune trees by @hinthornw in https://github.com/langchain-ai/langsmith-sdk/pull/1429

Full Changelog: https://github.com/langchain-ai/langsmith-sdk/compare/v0.2.11rc8...v0.2.11rc11

What's Changed

• Support parent=false to prune trees by @hinthornw in https://github.com/langchain-ai/langsmith-sdk/pull/1429

Full Changelog: https://github.com/langchain-ai/langsmith-sdk/compare/v0.2.11rc8...v0.2.11rc10

What's Changed

• Support parent=false to prune trees by @hinthornw in https://github.com/langchain-ai/langsmith-sdk/pull/1429

Full Changelog: https://github.com/langchain-ai/langsmith-sdk/compare/v0.2.11rc8...v0.2.11rc9

What's Changed

• feat(js): Add API refs build by @jacoblee93 in https://github.com/langchain-ai/langsmith-sdk/pull/1423
• Switch PyO3 to streaming zstd-compressed multipart Rust client. by @obi1kenobi in https://github.com/langchain-ai/langsmith-sdk/pull/1414
• Remove async and non-streaming blocking clients. by @obi1kenobi in https://github.com/langchain-ai/langsmith-sdk/pull/1419
• fix(js): Fix reflection in JS API refs by @jacoblee93 in https://github.com/langchain-ai/langsmith-sdk/pull/1426
• typo in benchmarking script by @isahers1 in https://github.com/langchain-ai/langsmith-sdk/pull/1427

Full Changelog: https://github.com/langchain-ai/langsmith-sdk/compare/v0.2.11rc7...v0.2.11rc8

What's Changed

• Serialize pydantic and other tricky objects correctly from Rust. by @obi1kenobi in https://github.com/langchain-ai/langsmith-sdk/pull/1392
• Implement streaming multipart zstd compression. by @obi1kenobi in https://github.com/langchain-ai/langsmith-sdk/pull/1413
• Re-vendor pyo3 at the v0.23.4 commit by @obi1kenobi in https://github.com/langchain-ai/langsmith-sdk/pull/1421

Full Changelog: https://github.com/langchain-ai/langsmith-sdk/compare/v0.2.11rc6...v0.2.11rc7

What's Changed

• surface mime type when passing attachments to user by @isahers1 in https://github.com/langchain-ai/langsmith-sdk/pull/1409

Full Changelog: https://github.com/langchain-ai/langsmith-sdk/compare/v0.2.11rc5...v0.2.11rc6

What's Changed

• python[patch]: update OpenAI prompt test by @baskaryan in https://github.com/langchain-ai/langsmith-sdk/pull/1407
• add mime type to attachment info by @isahers1 in https://github.com/langchain-ai/langsmith-sdk/pull/1341

Full Changelog: https://github.com/langchain-ai/langsmith-sdk/compare/v0.2.11rc4...v0.2.11rc5

What's Changed

• Make pydantic model serialization consistent regardless of surrogates. by @obi1kenobi in https://github.com/langchain-ai/langsmith-sdk/pull/1405

Full Changelog: https://github.com/langchain-ai/langsmith-sdk/compare/v0.2.11rc3...v0.2.11rc4

What's Changed

• Add safer accessors to delta objects by @hinthornw in https://github.com/langchain-ai/langsmith-sdk/pull/1385
• fix(python): Fix flaky openai parse tests by @angus-langchain in https://github.com/langchain-ai/langsmith-sdk/pull/1400
• Include user info in read feedback by @hinthornw in https://github.com/langchain-ai/langsmith-sdk/pull/1386
• make LANGSMITH_PROJECT work by @isahers1 in https://github.com/langchain-ai/langsmith-sdk/pull/1404

Full Changelog: https://github.com/langchain-ai/langsmith-sdk/compare/v0.2.11rc2...v0.2.11rc3

What's Changed

• fix(vercel): nested runs being unflattened, incorrect inheritance behavior of user runId / traceable by @dqbd in https://github.com/langchain-ai/langsmith-sdk/pull/1383
• release(js): 0.2.15 by @jacoblee93 in https://github.com/langchain-ai/langsmith-sdk/pull/1382
• docs: update api ref deps by @baskaryan in https://github.com/langchain-ai/langsmith-sdk/pull/1393

Full Changelog: https://github.com/langchain-ai/langsmith-sdk/compare/v0.2.11rc1...v0.2.11rc2

What's Changed

• feat(js): Add method to get size of annotation queue by @tendev-liam in https://github.com/langchain-ai/langsmith-sdk/pull/1377
• feat(js): Add method to delete a run from an annotation queue by @tendev-liam in https://github.com/langchain-ai/langsmith-sdk/pull/1376
• infra: allow non-main py release by @baskaryan in https://github.com/langchain-ai/langsmith-sdk/pull/1388
• Delete examples bulk by @hinthornw in https://github.com/langchain-ai/langsmith-sdk/pull/1279
• infra: fix py release inputs by @baskaryan in https://github.com/langchain-ai/langsmith-sdk/pull/1389

New Contributors

• @tendev-liam made their first contribution in https://github.com/langchain-ai/langsmith-sdk/pull/1377

Full Changelog: https://github.com/langchain-ai/langsmith-sdk/compare/v0.2.10...v0.2.11rc1

What's Changed

• feat: Instrument wrapper for OpenAI structured outputs by @angus-langchain in https://github.com/langchain-ai/langsmith-sdk/pull/1373
• chore: Version bump by @angus-langchain in https://github.com/langchain-ai/langsmith-sdk/pull/1374

Full Changelog: https://github.com/langchain-ai/langsmith-sdk/compare/v0.2.9...v0.2.10

What's Changed

• python[patch]: handle deployments without attachments support by @baskaryan in https://github.com/langchain-ai/langsmith-sdk/pull/1371
• python[patch]: Release 0.2.9 by @baskaryan in https://github.com/langchain-ai/langsmith-sdk/pull/1372

Full Changelog: https://github.com/langchain-ai/langsmith-sdk/compare/v0.2.8...v0.2.9

What's Changed

• feat: Make size_limit (max run count per batch) globally configurable via env var by @angus-langchain in https://github.com/langchain-ai/langsmith-sdk/pull/1358
• python[patch]: fix traceable arg inference by @baskaryan in https://github.com/langchain-ai/langsmith-sdk/pull/1361
• python[patch]: write to same experiment pytest-xdist by @baskaryan in https://github.com/langchain-ai/langsmith-sdk/pull/1367
• feat: Report pre and post compressed content length to API by @angus-langchain in https://github.com/langchain-ai/langsmith-sdk/pull/1359
• chore: Version bump by @angus-langchain in https://github.com/langchain-ai/langsmith-sdk/pull/1370

Full Changelog: https://github.com/langchain-ai/langsmith-sdk/compare/v0.2.7...v0.2.8

What's Changed

• fix: Separate rust and compressed ingestion paths by @angus-langchain in https://github.com/langchain-ai/langsmith-sdk/pull/1355
• fix: Flush zstd and correctly get buffer size by @angus-langchain in https://github.com/langchain-ai/langsmith-sdk/pull/1353
• fix: Reset buffer after sending to server by @angus-langchain in https://github.com/langchain-ai/langsmith-sdk/pull/1363
• chore: Version bump by @angus-langchain in https://github.com/langchain-ai/langsmith-sdk/pull/1364

Full Changelog: https://github.com/langchain-ai/langsmith-sdk/compare/v0.2.6...v0.2.7

What's Changed

• docs: update ref for list_examples by @isahers1 in https://github.com/langchain-ai/langsmith-sdk/pull/1346
• feat(js): Adds manual flush mode by @jacoblee93 in https://github.com/langchain-ai/langsmith-sdk/pull/1351
• chore: bump to 0.2.6 by @agola11 in https://github.com/langchain-ai/langsmith-sdk/pull/1352

Full Changelog: https://github.com/langchain-ai/langsmith-sdk/compare/v0.2.5...v0.2.6

What's Changed

• Improve the error message for tag and version mismatch error case. by @obi1kenobi in https://github.com/langchain-ai/langsmith-sdk/pull/1308
• Ensure pre-formatted JSON data doesn't contain trailing null bytes. by @obi1kenobi in https://github.com/langchain-ai/langsmith-sdk/pull/1332
• Merge Update operations into Create operations in the same batch. by @obi1kenobi in https://github.com/langchain-ai/langsmith-sdk/pull/1333
• Drain the Rust client when its last copy in Python is dropped. by @obi1kenobi in https://github.com/langchain-ai/langsmith-sdk/pull/1334
• chore: release langsmith-pyo3 version 0.1.0rc5 by @agola11 in https://github.com/langchain-ai/langsmith-sdk/pull/1335
• fix time for runs in convert_runs_test by @isahers1 in https://github.com/langchain-ai/langsmith-sdk/pull/1338
• only pop graph if 'serialized' is not none by @isahers1 in https://github.com/langchain-ai/langsmith-sdk/pull/1336
• python[patch]: evaluate accept list target by @baskaryan in https://github.com/langchain-ai/langsmith-sdk/pull/1342
• chore: version bump by @isahers1 in https://github.com/langchain-ai/langsmith-sdk/pull/1344

Full Changelog: https://github.com/langchain-ai/langsmith-sdk/compare/v0.2.3...v0.2.4

What's Changed

• fix: multiple evaluators don't consume the reader by @isahers1 in https://github.com/langchain-ai/langsmith-sdk/pull/1322
• feat(js): upload and update examples multipart in JS by @isahers1 in https://github.com/langchain-ai/langsmith-sdk/pull/1216
• chore: release JS SDK 0.2.12 by @agola11 in https://github.com/langchain-ai/langsmith-sdk/pull/1323
• remove attachment prefix by @isahers1 in https://github.com/langchain-ai/langsmith-sdk/pull/1324
• evaluate small fix by @isahers1 in https://github.com/langchain-ai/langsmith-sdk/pull/1327
• fix: allow passing object instead of tuples in JS since tuples are not nice in JS by @isahers1 in https://github.com/langchain-ai/langsmith-sdk/pull/1328
• chore: release python 0.2.3, js 0.2.13 by @agola11 in https://github.com/langchain-ai/langsmith-sdk/pull/1326

Full Changelog: https://github.com/langchain-ai/langsmith-sdk/compare/v0.2.2...v0.2.3

What's Changed

• Fix formatting in Rust tracing client. by @obi1kenobi in https://github.com/langchain-ai/langsmith-sdk/pull/1296
• Release langsmith-pyo3 version 0.1.0rc3. by @obi1kenobi in https://github.com/langchain-ai/langsmith-sdk/pull/1309
• Add Rust implementation for Python client's update_run() method. by @obi1kenobi in https://github.com/langchain-ai/langsmith-sdk/pull/1314
• chore: release langsmith-pyo3 version 0.1.0rc4 by @agola11 in https://github.com/langchain-ai/langsmith-sdk/pull/1315
• [python] support uploading examples with attachments and running evals on examples with attachments by @isahers1 in https://github.com/langchain-ai/langsmith-sdk/pull/1209

Full Changelog: https://github.com/langchain-ai/langsmith-sdk/compare/v0.2.1...v0.2.2

What's Changed

• chore: increase server info request timeout by @bvs-langchain in https://github.com/langchain-ai/langsmith-sdk/pull/1294
• Upgrade thiserror from v1 to v2. by @obi1kenobi in https://github.com/langchain-ai/langsmith-sdk/pull/1287
• chore: bump version to 0.2.10 to increase server info request timeout by @bvs-langchain in https://github.com/langchain-ai/langsmith-sdk/pull/1295
• fix(ci): Fix CI semver check by @jacoblee93 in https://github.com/langchain-ai/langsmith-sdk/pull/1297
• Rm hub pull check by @hinthornw in https://github.com/langchain-ai/langsmith-sdk/pull/1298
• release py=0.2.1, js=0.2.11 by @hinthornw in https://github.com/langchain-ai/langsmith-sdk/pull/1302

Full Changelog: https://github.com/langchain-ai/langsmith-sdk/compare/v0.2.0...v0.2.1

What's Changed

• feat(js): Expose light interface for tracing client by @jacoblee93 in https://github.com/langchain-ai/langsmith-sdk/pull/1266
• infra: drop py3.8 CI by @baskaryan in https://github.com/langchain-ai/langsmith-sdk/pull/1188
• python[patch]: one evaluate fn by @baskaryan in https://github.com/langchain-ai/langsmith-sdk/pull/1269
• python[patch]: client.evaluate, client.aevaluate by @baskaryan in https://github.com/langchain-ai/langsmith-sdk/pull/1268
• infra: run integration tests when relevant files edited by @baskaryan in https://github.com/langchain-ai/langsmith-sdk/pull/1271
• infra: fix int ci by @baskaryan in https://github.com/langchain-ai/langsmith-sdk/pull/1274
• infra: fix int ci 2 by @baskaryan in https://github.com/langchain-ai/langsmith-sdk/pull/1275
• js[patch]: simple evaluator args by @baskaryan in https://github.com/langchain-ai/langsmith-sdk/pull/1264
• Bagatur/update py api ref by @baskaryan in https://github.com/langchain-ai/langsmith-sdk/pull/1251
• js[patch]: fix evaluateComparative by @baskaryan in https://github.com/langchain-ai/langsmith-sdk/pull/1283
• js[patch]: evaluate support comparison by @baskaryan in https://github.com/langchain-ai/langsmith-sdk/pull/1281
• Support serializing Python UUID objects and set up test suite for PyO3 by @obi1kenobi in https://github.com/langchain-ai/langsmith-sdk/pull/1284
• Remove sonic_rs and switch to serde_json. by @obi1kenobi in https://github.com/langchain-ai/langsmith-sdk/pull/1286
• langsmith no longer supports Python 3.8, so neither should our PyO3 code by @obi1kenobi in https://github.com/langchain-ai/langsmith-sdk/pull/1285
• Drop outdated/vulnerable dependencies by using a newer multipart library by @obi1kenobi in https://github.com/langchain-ai/langsmith-sdk/pull/1288
• Run tests and lint for Rust code in CI. by @obi1kenobi in https://github.com/langchain-ai/langsmith-sdk/pull/1289
• Add placeholder workflow for testing langsmith_nodejs Rust code. by @obi1kenobi in https://github.com/langchain-ai/langsmith-sdk/pull/1291
• Clone dataset schema by @hinthornw in https://github.com/langchain-ai/langsmith-sdk/pull/1278
• LangSmith 0.2.x by @jakerachleff in https://github.com/langchain-ai/langsmith-sdk/pull/1247

Full Changelog: https://github.com/langchain-ai/langsmith-sdk/compare/v0.1.147...v0.2.0

What's Changed

• Ankush/pyo3 by @agola11 in https://github.com/langchain-ai/langsmith-sdk/pull/1139
• Release candidate langsmith-pyo3 Python package by @obi1kenobi in https://github.com/langchain-ai/langsmith-sdk/pull/1254
• feat: cross platform tracing client written in rust by @agola11 in https://github.com/langchain-ai/langsmith-sdk/pull/1107
• [perf] improve benchmark perf by @agola11 in https://github.com/langchain-ai/langsmith-sdk/pull/1121
• Add macos-15 runner to the PyO3 wheel build matrix. by @obi1kenobi in https://github.com/langchain-ai/langsmith-sdk/pull/1255
• Add Python 3.13 to matrix for building langsmith-pyo3 wheels. by @obi1kenobi in https://github.com/langchain-ai/langsmith-sdk/pull/1257
• Rename the macos artifacts to include the macOS version names. by @obi1kenobi in https://github.com/langchain-ai/langsmith-sdk/pull/1258
• Build 3.13 wheels for all platforms except Windows. by @obi1kenobi in https://github.com/langchain-ai/langsmith-sdk/pull/1261
• Remove links in vendored code that point outside the sdist source code. by @obi1kenobi in https://github.com/langchain-ai/langsmith-sdk/pull/1256
• Ensure the wheel-building workflow runs if it's changed. by @obi1kenobi in https://github.com/langchain-ai/langsmith-sdk/pull/1259
• Make langsmith-pyo3 v0.1.0rc2 release. by @obi1kenobi in https://github.com/langchain-ai/langsmith-sdk/pull/1260
• Python wheels don't distinguish between macOS versions. by @obi1kenobi in https://github.com/langchain-ai/langsmith-sdk/pull/1262
• Add a langsmith_pyo3 extra to the LangSmith SDK. by @obi1kenobi in https://github.com/langchain-ai/langsmith-sdk/pull/1263
• chore: bump version to 0.1.147 by @agola11 in https://github.com/langchain-ai/langsmith-sdk/pull/1265

Full Changelog: https://github.com/langchain-ai/langsmith-sdk/compare/v0.1.146...v0.1.147

What's Changed

• Improve support for pulling structured prompts w/ model info by @hinthornw in https://github.com/langchain-ai/langsmith-sdk/pull/1210

Full Changelog: https://github.com/langchain-ai/langsmith-sdk/compare/v0.1.145...v0.1.146

What's Changed

• Attachments in patch by @hinthornw in https://github.com/langchain-ai/langsmith-sdk/pull/1212
• fix(js): Add LANGSMITH_ env vars to metadata by @jacoblee93 in https://github.com/langchain-ai/langsmith-sdk/pull/1243
• Add placeholder workflow for building PyO3 wheels. by @obi1kenobi in https://github.com/langchain-ai/langsmith-sdk/pull/1244
• python[patch]: summary evaluator simpler returns by @baskaryan in https://github.com/langchain-ai/langsmith-sdk/pull/1241
• python[patch]: comparison evaluator simplification by @baskaryan in https://github.com/langchain-ai/langsmith-sdk/pull/1240
• recompile pattern if needed before its used to mask sensitive info by @davidx33 in https://github.com/langchain-ai/langsmith-sdk/pull/1248
• python[patch]: evaluate local mode by @baskaryan in https://github.com/langchain-ai/langsmith-sdk/pull/1224
• Refrain from overwriting experiment metadata by @hinthornw in https://github.com/langchain-ai/langsmith-sdk/pull/1250

New Contributors

• @davidx33 made their first contribution in https://github.com/langchain-ai/langsmith-sdk/pull/1248

Full Changelog: https://github.com/langchain-ai/langsmith-sdk/compare/v0.1.144...v0.1.145

What's Changed

• Lazy internal import by @hinthornw in https://github.com/langchain-ai/langsmith-sdk/pull/1198
• python[patch]: evaluators can return primitives by @baskaryan in https://github.com/langchain-ai/langsmith-sdk/pull/1203
• perf: cut down cpu time of aevaluate by 30% on 1-4MB examples with this one trick by @jakerachleff in https://github.com/langchain-ai/langsmith-sdk/pull/1217
• nit: stop calling importlib on every api req by @jakerachleff in https://github.com/langchain-ai/langsmith-sdk/pull/1219
• python[patch]: evaluate accept Dataset by @baskaryan in https://github.com/langchain-ai/langsmith-sdk/pull/1222
• python[patch]: accept simple evaluators by @baskaryan in https://github.com/langchain-ai/langsmith-sdk/pull/1200
• NotMyPy by @hinthornw in https://github.com/langchain-ai/langsmith-sdk/pull/1221
• orjson optional wip by @efriis in https://github.com/langchain-ai/langsmith-sdk/pull/1223
• release 0.1.144rc1 by @efriis in https://github.com/langchain-ai/langsmith-sdk/pull/1226
• Revert "orjson optional wip (#1223)" by @baskaryan in https://github.com/langchain-ai/langsmith-sdk/pull/1228
• orjson optional take 2 by @efriis in https://github.com/langchain-ai/langsmith-sdk/pull/1230
• skip flaky js integration test by @efriis in https://github.com/langchain-ai/langsmith-sdk/pull/1232
• infra: use benchmark-fast in pybench ci by @efriis in https://github.com/langchain-ai/langsmith-sdk/pull/1160
• timeout_ms docstring by @efriis in https://github.com/langchain-ai/langsmith-sdk/pull/1234
• fix(js): Fix multipart upload with overridden node fetch by @jacoblee93 in https://github.com/langchain-ai/langsmith-sdk/pull/1235
• python[patch]: accept simple summary eval signature by @baskaryan in https://github.com/langchain-ai/langsmith-sdk/pull/1227
• Bump cross-spawn from 7.0.3 to 7.0.6 in /js by @dependabot in https://github.com/langchain-ai/langsmith-sdk/pull/1233
• sdk patch: raise errors on update/create examples when passing mismatching length sequences by @isahers1 in https://github.com/langchain-ai/langsmith-sdk/pull/1238
• Support comments in evaluate_comparative by @hinthornw in https://github.com/langchain-ai/langsmith-sdk/pull/1211
• [python] Ensure timezone by @hinthornw in https://github.com/langchain-ai/langsmith-sdk/pull/1239

Full Changelog: https://github.com/langchain-ai/langsmith-sdk/compare/v0.1.143...v0.1.144

What's Changed

• Lazy internal import by @hinthornw in https://github.com/langchain-ai/langsmith-sdk/pull/1198
• python[patch]: evaluators can return primitives by @baskaryan in https://github.com/langchain-ai/langsmith-sdk/pull/1203
• perf: cut down cpu time of aevaluate by 30% on 1-4MB examples with this one trick by @jakerachleff in https://github.com/langchain-ai/langsmith-sdk/pull/1217
• nit: stop calling importlib on every api req by @jakerachleff in https://github.com/langchain-ai/langsmith-sdk/pull/1219
• python[patch]: evaluate accept Dataset by @baskaryan in https://github.com/langchain-ai/langsmith-sdk/pull/1222
• python[patch]: accept simple evaluators by @baskaryan in https://github.com/langchain-ai/langsmith-sdk/pull/1200
• NotMyPy by @hinthornw in https://github.com/langchain-ai/langsmith-sdk/pull/1221
• orjson optional wip by @efriis in https://github.com/langchain-ai/langsmith-sdk/pull/1223
• release 0.1.144rc1 by @efriis in https://github.com/langchain-ai/langsmith-sdk/pull/1226

Full Changelog: https://github.com/langchain-ai/langsmith-sdk/compare/v0.1.143...v0.1.144rc1

What's Changed

• [Python] push_prompt update by @hinthornw in https://github.com/langchain-ai/langsmith-sdk/pull/1202
• python[patch]: pass Runnable to evaluate by @baskaryan in https://github.com/langchain-ai/langsmith-sdk/pull/1204

Full Changelog: https://github.com/langchain-ai/langsmith-sdk/compare/v0.1.142...v0.1.143

What's Changed

• fix(js): Use batch endpoint instead of individual run endpoint in case of server info fetch failure by @jacoblee93 in https://github.com/langchain-ai/langsmith-sdk/pull/1189
• Fix urllib patch path by @hinthornw in https://github.com/langchain-ai/langsmith-sdk/pull/1197

Full Changelog: https://github.com/langchain-ai/langsmith-sdk/compare/v0.1.141...v0.1.142

What's Changed

• js feat: Support attachments in traceable by @hinthornw in https://github.com/langchain-ai/langsmith-sdk/pull/1169
• Improve error msg for empty data by @hinthornw in https://github.com/langchain-ai/langsmith-sdk/pull/1178
• chore: update fastapi dev dependency to fix vulnerability report by @agola11 in https://github.com/langchain-ai/langsmith-sdk/pull/1182
• [Python]: Fix concurrency issues in caching for urllib3<2.0 by @hinthornw in https://github.com/langchain-ai/langsmith-sdk/pull/1183
• Remove initial timeout, the batch timeout is now always 250ms by @nhuang-lc in https://github.com/langchain-ai/langsmith-sdk/pull/1184
• fix(js): Export more evaluator types by @bracesproul in https://github.com/langchain-ai/langsmith-sdk/pull/1190
• fix(js): Release 0.2.5 by @bracesproul in https://github.com/langchain-ai/langsmith-sdk/pull/1191
• [Python] Update prompt schema by @hinthornw in https://github.com/langchain-ai/langsmith-sdk/pull/1193
• Python 0.1.141 by @hinthornw in https://github.com/langchain-ai/langsmith-sdk/pull/1194

Full Changelog: https://github.com/langchain-ai/langsmith-sdk/compare/v0.1.140...v0.1.141

What's Changed

• track evaluator errors from sdk by @isahers1 in https://github.com/langchain-ai/langsmith-sdk/pull/1079
• feat(vercel): add debug flag by @dqbd in https://github.com/langchain-ai/langsmith-sdk/pull/1166
• fix(vercel): add more debug logs, fix tracing detection for vercel, bump to 0.2.4 by @dqbd in https://github.com/langchain-ai/langsmith-sdk/pull/1168
• Fix clone dataset from other server by @hinthornw in https://github.com/langchain-ai/langsmith-sdk/pull/1175
• Cleaner error message by @hinthornw in https://github.com/langchain-ai/langsmith-sdk/pull/1177
• [Python] to_pandas() method for ExperimentResults by @baskaryan in https://github.com/langchain-ai/langsmith-sdk/pull/1170
• Pre-read smallish multipart requests by @hinthornw in https://github.com/langchain-ai/langsmith-sdk/pull/1176

New Contributors

• @isahers1 made their first contribution in https://github.com/langchain-ai/langsmith-sdk/pull/1079

Full Changelog: https://github.com/langchain-ai/langsmith-sdk/compare/v0.1.139...v0.1.140

What's Changed

• Skip model_dump for classes by @hinthornw in https://github.com/langchain-ai/langsmith-sdk/pull/1149
• Minimize Repr by @hinthornw in https://github.com/langchain-ai/langsmith-sdk/pull/1155
• serialize instead of deepcopy for performance gains by @efriis in https://github.com/langchain-ai/langsmith-sdk/pull/1136
• Annotation Support by @hinthornw in https://github.com/langchain-ai/langsmith-sdk/pull/1146
• release 0.1.139 by @efriis in https://github.com/langchain-ai/langsmith-sdk/pull/1161

Full Changelog: https://github.com/langchain-ai/langsmith-sdk/compare/v0.1.138...v0.1.139

What's Changed

• Annotation Support by @hinthornw in https://github.com/langchain-ai/langsmith-sdk/pull/1146

Full Changelog: https://github.com/langchain-ai/langsmith-sdk/compare/v0.1.139rc1...v0.1.139rc2

What's Changed

• Skip model_dump for classes by @hinthornw in https://github.com/langchain-ai/langsmith-sdk/pull/1149
• Minimize Repr by @hinthornw in https://github.com/langchain-ai/langsmith-sdk/pull/1155
• serialize instead of deepcopy for performance gains by @efriis in https://github.com/langchain-ai/langsmith-sdk/pull/1136

Full Changelog: https://github.com/langchain-ai/langsmith-sdk/compare/v0.1.138...v0.1.139rc1

What's Changed

• feat(js): Use new multipart endpoint when available by @jacoblee93 in https://github.com/langchain-ai/langsmith-sdk/pull/1106
• chore(js): Release 0.1.67 by @jacoblee93 in https://github.com/langchain-ai/langsmith-sdk/pull/1119
• fix(js): Limit queue batch concurrency, bump timeout, add maximum wait period for serverless envs by @jacoblee93 in https://github.com/langchain-ai/langsmith-sdk/pull/1124
• feat(vercel): add OTEL based LangSmith trace exporter by @dqbd in https://github.com/langchain-ai/langsmith-sdk/pull/1104
• factor out background and serde by @efriis in https://github.com/langchain-ai/langsmith-sdk/pull/1130
• integration tests multipart by @efriis in https://github.com/langchain-ai/langsmith-sdk/pull/1118
• properly setup env for tests by @efriis in https://github.com/langchain-ai/langsmith-sdk/pull/1131
• fix(js): Further reduce blocking behavior, try naive stringification instead of circular detection by @jacoblee93 in https://github.com/langchain-ai/langsmith-sdk/pull/1133
• integration tests against staging by @efriis in https://github.com/langchain-ai/langsmith-sdk/pull/1143
• fix(js): Remove focused test, fix tests by @jacoblee93 in https://github.com/langchain-ai/langsmith-sdk/pull/1138
• fix: add required basic auth environment variable by @bvs-langchain in https://github.com/langchain-ai/langsmith-sdk/pull/1128
• feat(py): Add multipart feedback ingestion by @akira in https://github.com/langchain-ai/langsmith-sdk/pull/1129
• use staging for integration tests - js by @efriis in https://github.com/langchain-ai/langsmith-sdk/pull/1145
• use staging for integration tests by @efriis in https://github.com/langchain-ai/langsmith-sdk/pull/1140
• fix(vercel): pass run name if the root run is within a span by @dqbd in https://github.com/langchain-ai/langsmith-sdk/pull/1150
• Add version check for feedback multipart by @akira in https://github.com/langchain-ai/langsmith-sdk/pull/1151
• [python] Bump to 0.1.138 by @akira in https://github.com/langchain-ai/langsmith-sdk/pull/1156

New Contributors

• @akira made their first contribution in https://github.com/langchain-ai/langsmith-sdk/pull/1129

Full Changelog: https://github.com/langchain-ai/langsmith-sdk/compare/v0.1.137...v0.1.138

What's Changed

• integration tests against staging by @efriis in https://github.com/langchain-ai/langsmith-sdk/pull/1143
• use staging for integration tests - js by @efriis in https://github.com/langchain-ai/langsmith-sdk/pull/1145
• use staging for integration tests by @efriis in https://github.com/langchain-ai/langsmith-sdk/pull/1140
• fix(vercel): pass run name if the root run is within a span by @dqbd in https://github.com/langchain-ai/langsmith-sdk/pull/1150
• Add version check for feedback multipart by @akira in https://github.com/langchain-ai/langsmith-sdk/pull/1151

Full Changelog: https://github.com/langchain-ai/langsmith-sdk/compare/v0.1.138rc1...v0.1.138rc2

What's Changed

• feat(js): Use new multipart endpoint when available by @jacoblee93 in https://github.com/langchain-ai/langsmith-sdk/pull/1106
• chore(js): Release 0.1.67 by @jacoblee93 in https://github.com/langchain-ai/langsmith-sdk/pull/1119
• fix(js): Limit queue batch concurrency, bump timeout, add maximum wait period for serverless envs by @jacoblee93 in https://github.com/langchain-ai/langsmith-sdk/pull/1124
• feat(vercel): add OTEL based LangSmith trace exporter by @dqbd in https://github.com/langchain-ai/langsmith-sdk/pull/1104
• factor out background and serde by @efriis in https://github.com/langchain-ai/langsmith-sdk/pull/1130
• integration tests multipart by @efriis in https://github.com/langchain-ai/langsmith-sdk/pull/1118
• properly setup env for tests by @efriis in https://github.com/langchain-ai/langsmith-sdk/pull/1131
• fix(js): Further reduce blocking behavior, try naive stringification instead of circular detection by @jacoblee93 in https://github.com/langchain-ai/langsmith-sdk/pull/1133
• fix(js): Remove focused test, fix tests by @jacoblee93 in https://github.com/langchain-ai/langsmith-sdk/pull/1138
• fix: add required basic auth environment variable by @bvs-langchain in https://github.com/langchain-ai/langsmith-sdk/pull/1128
• feat(py): Add multipart feedback ingestion by @akira in https://github.com/langchain-ai/langsmith-sdk/pull/1129

New Contributors

• @akira made their first contribution in https://github.com/langchain-ai/langsmith-sdk/pull/1129

Full Changelog: https://github.com/langchain-ai/langsmith-sdk/compare/v0.1.137...v0.1.138rc1

What's Changed

• fix: Create new multipart encoder for each retry attempt by @nfcampos in https://github.com/langchain-ai/langsmith-sdk/pull/1114

Full Changelog: https://github.com/langchain-ai/langsmith-sdk/compare/v0.1.136...v0.1.137

What's Changed

• feat(JS): log wrapOpenAI calls with usage metdata by @agola11 in https://github.com/langchain-ai/langsmith-sdk/pull/1096
• feat: log wrap_openai runs with unified usage_metadata by @agola11 in https://github.com/langchain-ai/langsmith-sdk/pull/1071
• feat(JS): add processInputs / processOutputs to traceable + add usage metadata to wrapOpenAI by @agola11 in https://github.com/langchain-ai/langsmith-sdk/pull/1095
• bump version by @nhuang-lc in https://github.com/langchain-ai/langsmith-sdk/pull/1100
• py: Remove attachments from main run payload before upload by @nfcampos in https://github.com/langchain-ai/langsmith-sdk/pull/1103
• py: 0.1.136 by @nfcampos in https://github.com/langchain-ai/langsmith-sdk/pull/1105

Full Changelog: https://github.com/langchain-ai/langsmith-sdk/compare/v0.1.135...v0.1.136

What's Changed

• feat(js): Validate that store values are RunTree instances by @jacoblee93 in https://github.com/langchain-ai/langsmith-sdk/pull/1088
• chore(js): Release 0.1.64 by @jacoblee93 in https://github.com/langchain-ai/langsmith-sdk/pull/1092
• Allow users to add metadata on end() by @nhuang-lc in https://github.com/langchain-ai/langsmith-sdk/pull/1085
• Add UUID to project name in integration test by @nhuang-lc in https://github.com/langchain-ai/langsmith-sdk/pull/1094
• feat(js): Adds further required support for context vars by @jacoblee93 in https://github.com/langchain-ai/langsmith-sdk/pull/1093
• py: Enable multipart endpoint without feature flag by @nfcampos in https://github.com/langchain-ai/langsmith-sdk/pull/1098
• py 0.1.135 by @nfcampos in https://github.com/langchain-ai/langsmith-sdk/pull/1099

Full Changelog: https://github.com/langchain-ai/langsmith-sdk/compare/v0.1.134...v0.1.135

What's Changed

• py: Add session_id and session_name to patch parts in batch/multipart by @nfcampos in https://github.com/langchain-ai/langsmith-sdk/pull/1086
• Py 0.1.134 by @nfcampos in https://github.com/langchain-ai/langsmith-sdk/pull/1087

Full Changelog: https://github.com/langchain-ai/langsmith-sdk/compare/v0.1.133...v0.1.134

What's Changed

• Update README.md by @eyurtsev in https://github.com/langchain-ai/langsmith-sdk/pull/1072
• feat(js): Allow full backgrounding of auto batches by @jacoblee93 in https://github.com/langchain-ai/langsmith-sdk/pull/1069
• py: Retry on 408 by @nfcampos in https://github.com/langchain-ai/langsmith-sdk/pull/1075
• fix(js): More deeply copy OpenAI SDK when wrapping by @jacoblee93 in https://github.com/langchain-ai/langsmith-sdk/pull/1081
• Add distributed tracing middleware by @hinthornw in https://github.com/langchain-ai/langsmith-sdk/pull/1083
• Better async error message by @hinthornw in https://github.com/langchain-ai/langsmith-sdk/pull/1080
• Send collected outputs for generators by @hinthornw in https://github.com/langchain-ai/langsmith-sdk/pull/1082

Full Changelog: https://github.com/langchain-ai/langsmith-sdk/compare/v0.1.132...v0.1.133

What's Changed

• py: For multipart endpoint don't split into batches of 20mb by @nfcampos in https://github.com/langchain-ai/langsmith-sdk/pull/1067
• [Py] 0.1.132 by @hinthornw in https://github.com/langchain-ai/langsmith-sdk/pull/1070

Full Changelog: https://github.com/langchain-ai/langsmith-sdk/compare/v0.1.131...v0.1.132

What's Changed

• Run baseline benchmark job whenever bench code changes by @nfcampos in https://github.com/langchain-ai/langsmith-sdk/pull/1058
• py: Stop sending serialized as a separate part by @nfcampos in https://github.com/langchain-ai/langsmith-sdk/pull/1061
• [Py]Better mock handling in serialization by @hinthornw in https://github.com/langchain-ai/langsmith-sdk/pull/1063

Full Changelog: https://github.com/langchain-ai/langsmith-sdk/compare/v0.1.130...v0.1.131

What's Changed

• chore: bump sdk 0.7.39 by @langchain-infra in https://github.com/langchain-ai/langsmith-sdk/pull/1047
• fix: remove password req by @langchain-infra in https://github.com/langchain-ai/langsmith-sdk/pull/1050
• Setup Bench by @hinthornw in https://github.com/langchain-ai/langsmith-sdk/pull/1048
• fix: add docker-compose tests by @langchain-infra in https://github.com/langchain-ai/langsmith-sdk/pull/1051
• Faster Serialization by @hinthornw in https://github.com/langchain-ai/langsmith-sdk/pull/984
• chore: update docker-compose for new OAuth mode by @bvs-langchain in https://github.com/langchain-ai/langsmith-sdk/pull/1052
• chore: add example env vars for OAuth2.0 with client secret by @bvs-langchain in https://github.com/langchain-ai/langsmith-sdk/pull/1053
• Implement "streaming" multipart requests to multipart endpoint by @nfcampos in https://github.com/langchain-ai/langsmith-sdk/pull/1055
• Nc/1oct/py prepare multipart by @nfcampos in https://github.com/langchain-ai/langsmith-sdk/pull/1054
• Fix benchmarks by @nfcampos in https://github.com/langchain-ai/langsmith-sdk/pull/1056
• [Python] 0.1.130 by @hinthornw in https://github.com/langchain-ai/langsmith-sdk/pull/1060

Full Changelog: https://github.com/langchain-ai/langsmith-sdk/compare/v0.1.129...v0.1.130

What's Changed

• [JS] Catch any errors in create/update run by @hinthornw in https://github.com/langchain-ai/langsmith-sdk/pull/1044
• Handle missing version errors by @hinthornw in https://github.com/langchain-ai/langsmith-sdk/pull/1046

Full Changelog: https://github.com/langchain-ai/langsmith-sdk/compare/v0.1.128...v0.1.129

What's Changed

• [Python] increase default client timeout in evals by @hinthornw in https://github.com/langchain-ai/langsmith-sdk/pull/1030
• [Python] List prompts by @hinthornw in https://github.com/langchain-ai/langsmith-sdk/pull/1039

Full Changelog: https://github.com/langchain-ai/langsmith-sdk/compare/v0.1.127...v0.1.128

What's Changed

• fix(js): Trim trailing slashes on passed API and web URLs by @jacoblee93 in https://github.com/langchain-ai/langsmith-sdk/pull/1033
• fix(js): Log instead of throw errors on 404 by @jacoblee93 in https://github.com/langchain-ai/langsmith-sdk/pull/1032
• chore(js): Release 0.1.60 by @jacoblee93 in https://github.com/langchain-ai/langsmith-sdk/pull/1034
• [Python] Support stream responses by @hinthornw in https://github.com/langchain-ai/langsmith-sdk/pull/1028

Full Changelog: https://github.com/langchain-ai/langsmith-sdk/compare/v0.1.126...v0.1.127

What's Changed

• Adding method for removing a run from an annotation queue (#1027) by @hinthornw in https://github.com/langchain-ai/langsmith-sdk/pull/1029
• python[patch]: exclude RunTree._client by @baskaryan in https://github.com/langchain-ai/langsmith-sdk/pull/1031

Full Changelog: https://github.com/langchain-ai/langsmith-sdk/compare/v0.1.125...v0.1.126

What's Changed

• [Py] Default to unnamed by @hinthornw in https://github.com/langchain-ai/langsmith-sdk/pull/1024

Full Changelog: https://github.com/langchain-ai/langsmith-sdk/compare/v0.1.124...v0.1.125

What's Changed

• Change the 'text' model type to 'llm' to match other libraries by @efriis in https://github.com/langchain-ai/langsmith-sdk/pull/1019
• [Py] Add default_factory in run tree by @hinthornw in https://github.com/langchain-ai/langsmith-sdk/pull/1023

Full Changelog: https://github.com/langchain-ai/langsmith-sdk/compare/v0.1.123...v0.1.124

What's Changed

• Fix two descriptions in python sdk by @nhuang-lc in https://github.com/langchain-ai/langsmith-sdk/pull/1016
• Fix SHORT_LIVED_TTL_SECONDS wrong comment on default value by @abozaralizadeh in https://github.com/langchain-ai/langsmith-sdk/pull/1009
• [Py] Defaults in Example by @hinthornw in https://github.com/langchain-ai/langsmith-sdk/pull/1017

New Contributors

• @abozaralizadeh made their first contribution in https://github.com/langchain-ai/langsmith-sdk/pull/1009

Full Changelog: https://github.com/langchain-ai/langsmith-sdk/compare/v0.1.122...v0.1.123

What's Changed

• fix(js): Removes langchain and @langchain/core from peer deps to fix npm installation by @jacoblee93 in https://github.com/langchain-ai/langsmith-sdk/pull/1005
• [JS] Share client across run trees by @hinthornw in https://github.com/langchain-ai/langsmith-sdk/pull/1011
• Add methods in slots attribute by @nhuang-lc in https://github.com/langchain-ai/langsmith-sdk/pull/1012
• [Python] Filter beta warning in prompt deserialiazation by @hinthornw in https://github.com/langchain-ai/langsmith-sdk/pull/1013
• Always trace evaluators by @hinthornw in https://github.com/langchain-ai/langsmith-sdk/pull/1014
• [JS] Anno crud by @hinthornw in https://github.com/langchain-ai/langsmith-sdk/pull/1010

New Contributors

• @nhuang-lc made their first contribution in https://github.com/langchain-ai/langsmith-sdk/pull/1012

Full Changelog: https://github.com/langchain-ai/langsmith-sdk/compare/v0.1.121...v0.1.122

What's Changed

• [Python] Use _name by @hinthornw in https://github.com/langchain-ai/langsmith-sdk/pull/1008

Full Changelog: https://github.com/langchain-ai/langsmith-sdk/compare/v0.1.120...v0.1.121

What's Changed

• Support evaluate on existing experiment by @hinthornw in https://github.com/langchain-ai/langsmith-sdk/pull/1000

Full Changelog: https://github.com/langchain-ai/langsmith-sdk/compare/v0.1.119...v0.1.120