Releases: langchain-ai/langchain-nvidia
Releases · langchain-ai/langchain-nvidia
libs/ai-endpoints/v0.3.3
What's Changed
- add new retriever models: nvidia/llama-3.2-nv-embedqa-1b-v1 and nvidia/llama-3.2-nv-rerankqa-1b-v1 by @mattf in #113
Full Changelog: libs/ai-endpoints/v0.3.2...libs/ai-endpoints/v0.3.3
libs/ai-endpoints/v0.3.2
What's Changed
- allow base_url to work with proxies, e.g. http://host/proxy/path/v1 by @mattf in #112
Full Changelog: libs/ai-endpoints/v0.3.1...libs/ai-endpoints/v0.3.2
libs/ai-endpoints/v0.3.1
What's Changed
Full Changelog: libs/ai-endpoints/v0.3.0...libs/ai-endpoints/v0.3.1
libs/ai-endpoints/v0.3.0
What's Changed
- add example to rerank docs by @aviolante in #96
- add hosted nim in docstring by @aviolante in #98
- fix missing auth header on polling requests by @mattf in #99
- feat: include usage metadata by @mattf in #94
- migrate to langchain-core 0.3 (langchain_core.pydantic_v1 removal) by @baskaryan in #101
- add support for vision-language model endpoints by @mattf in #100
- ai-endpoints[patch]: standard tests by @efriis in #18
- remove deprecated model_type param (embedding) by @mattf in #102
- remove compatibility mode for nvolveqa_40k embedding model by @mattf in #103
- add strict support by @mattf in #104
- allow
import langchain_nvidia
by @mattf in #105
New Contributors
- @aviolante made their first contribution in #96
- @baskaryan made their first contribution in #101
Full Changelog: libs/ai-endpoints/v0.2.2...libs/ai-endpoints/v0.3.0
libs/ai-endpoints/v0.2.2
What's Changed
- Quick improvement to langgraph example by @haydeniw in #87
- update default models by @mattf in #89
- Fix: Raise a warning for known endpoints by @raspawar in #93
- add Completions API support by @mattf in #88
- bug: fix stream collection by @mattf in #95
- fix failing test cases for base url by @raspawar in #97
Full Changelog: libs/ai-endpoints/v0.2.1...libs/ai-endpoints/v0.2.2
libs/ai-endpoints/v0.2.1
What's Changed
- Added tool calling notebook by @haydeniw in #77
- ai-endpoints: remove langchain-community dependency by @efriis in #78
- chat nb update with tool calling and upstream changes by @dglogo in #80
- update nvidia.mdx file for stale links by @dglogo in #83
- Validation for base_url by @raspawar in #68
- Enable base_url to read from env variables by @kylehh in #65
- add with_structured_output support for Pydantic models, dicts and Enum by @mattf in #79
- Client rejects incompatible models by @raspawar in #81
- various internal improvements by @mattf in #82
- support pydantic v2 and v1 for with_structured_output by @mattf in #84
- Update min version for langchain-core by @raspawar in #85
New Contributors
Full Changelog: libs/ai-endpoints/v0.2.0...libs/ai-endpoints/v0.2.1
libs/ai-endpoints/v0.2.0
What's Changed
Full Changelog: libs/ai-endpoints/v0.1.7...libs/ai-endpoints/v0.2.0
libs/ai-endpoints/v0.1.7
libs/ai-endpoints/v0.1.6
libs/ai-endpoints/v0.1.5
What's Changed
- update default chat model to meta/llama3-8b-instruct (most features & deployment options) by @mattf in #73
- add support for reranking api change w/ truncate parameter by @mattf in #75
Full Changelog: libs/ai-endpoints/v0.1.4...libs/ai-endpoints/v0.1.5