ollama[patch]: support ollama 0.4 #28364
Merged
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
|
The latest updates on your projects. Learn more about Vercel for Git ↗︎ 1 Skipped Deployment
|
dosubot
bot
added
the
size:XS
AlwaysBluer
pushed a commit
to AlwaysBluer/langchain
that referenced
this pull request
Nov 27, 2024
… into LindormIntegration * 'LindormIntegration' of github.com:AlwaysBluer/langchain: (72 commits) langchain[patch]: update deprecation message for agent classes and constructors (langchain-ai#28369) Updated docs for the BM25 preprocessing function (langchain-ai#28101) Community: Marqo Index Setting GET Request Updated according to `2.x` API version while keep backward compatability for 1.5.x (langchain-ai#28342) langchain: Fix Neo4jVector vector store reference from partner package for self query (langchain-ai#28292) ollama[patch]: bump to 0.4.1 in lock file (langchain-ai#28365) ollama[patch]: support ollama 0.4 (langchain-ai#28364) openai[patch]: bump core dep (langchain-ai#28361) openai[patch]: Release 0.2.10 (langchain-ai#28360) openai[patch]: use max_completion_tokens in place of max_tokens (langchain-ai#26917) langchain[patch]: support numpy 2 (langchain-ai#28183) docs: update tutorials (langchain-ai#28219) docs[patch]: fix broken anchor link (langchain-ai#28358) docs(create_sql_agent): fix reStructured Text Markup (langchain-ai#28356) Community: Adding citations in AIMessage for ChatPerplexity (langchain-ai#28321) docs: update azure_openai_whisper_parser.ipynb (langchain-ai#28327) community[patch]: fix import in test (langchain-ai#28339) community[patch]: Neo4j community deprecation (langchain-ai#28130) community: fixed critical bugs at Writer provider (langchain-ai#27879) docs: fix GOOGLE_API_KEY typo (langchain-ai#28322) community[patch]: release 0.3.8 (langchain-ai#28316) ...
|
@ccurme For sure. I encountered some weird behavior while trying to work with multimodal input to an Ollama model following this tutorial. I then switched to using chatollama, and encountered an error in the serialization of the model. At this point I was in contact with @efriis, who I believe worked with the ollama folks to debug and make the change in #28299. ollama version: Initial behavior following tutorial: Error serializing model: |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
v0.4 of the Python SDK is already installed via the lock file in CI, but our current implementation is not compatible with it.
This also addresses an issue introduced in #28299. @RyanMagnuson would you mind explaining the motivation for that change? From what I can tell the Ollama SDK does not support kwargs. Previously, unsupported kwargs were ignored, but they currently raise
TypeError.Some of LangChain's standard test suite expects
tool_choiceto be supported, so here we catch it inbind_toolsso it is ignored and not passed through to the client.