We don't publish
your competitive advantage.

IFSupport for Reasoning in OpenRouter and DeepSeek Providers

THEN- [ ] I believe this is a way to improve. I'll try to join the [Continue Discord](https://discord.gg/NWtdYexhMs) for questions - [ ] I'm not able to find an [open issue](https://github.com/continuedev/continue/issues?q=is%3Aopen+is%3Aissue+label%3Aenhancement) that requests the same enhancement

IFServer capabilities not affecting the stream of "{capability}/list" requests and "method not found" errors.

THEN**Describe the bug** If you review the mcp diagnostic logs of basically any server (at least all the ones I've seen) they are littered with requests for capability lists and then "method not found" errors. It seems like the mcp client should read the capbilities once and then not query the undefined ones every couple seconds.

IFPatrick von Platen

THENDear Patrick von Platen I have some questions about the use of Helsinki-NLP models in commercial projects, can you please contact me by email? alan.heirich@gmail.com Sorry to use this issue tracker, I know of no other way to reach Patrick.

IFRuntimeError: CUDA error: CUBLAS_STATUS_NOT_INITIALIZED when calling `cublasCreate(handle)`

THENHi, I tried to add some other embeddings in your BertEmbedding source code and then load the pretrained weights 'bert-base-chinese'. When I run the forward method, I got the issue 'RuntimeError: CUDA error: CUBLAS_STATUS_NOT_INITIALIZED when calling `cublasCreate(handle)`' Can someone help please? Thanks a lot

IF[BUG] Frequent IDE disconnections disrupting workflow in VS Code integration

THENClaude CLI diagnostic output: Claude CLI Diagnostic Currently running: unknown (1.0.64) Path: /Users/bryan/.proto/tools/node/22.17.1/bin/node Invoked: /Users/bryan/.npm-global/bin/claude Config install method: unknown Auto-updates enabled: true

IF[Bug]: Multiple MetaDataFilters not working even with a condition set.

THEN``` query_engine = index.as_query_engine( filters=MetadataFilters( filters=[ MetadataFilter( key="color", value="red", ), MetadataFilter( key="alpha", value="1", ), ], condition=FilterCondition.OR, ) ``` Only when i use either of the filters it

IFSupport download of models from Torrent

THENModels could be censored and removed from HF. In order to keep them alive, Torrent could be a solution.

IF[Bug]: ui.txt endpoint return 404

THENI see that on my instance I have this error on console:

IFautoCompletion: deepseek - Stop string array too long

THEN--- Hi there, We've increased the limit at our side, and the new limit for the stop string array is now 16. However, please note that the 236B coder v2 model does not support FIM yet, and the feature is still work-in-progress. I built a prompt to replace FIM special tokens, and maybe you can give it a try: ```json { "models": [], "tabAutocompleteOptions": { "template": "Please te

IF[Feature Request]: how do we add tools to site_1 created by AgentBuilder ?

THENI want to integrate my custom functions and langchain tools with site_1 created by AgentBuilder

IFChatOllama & Ollama from langchain_ollama partner package does not provide support to pass base_url

THEN- [X] I added a very descriptive title to this issue. - [X] I searched the LangChain documentation with the integrated search. - [X] I used the GitHub search to find a similar question and didn't find it. - [X] I am sure that this is a bug in LangChain rather than my code. - [X] The bug is not resolved by updating to the latest stable version of LangChain (or the specific integration package).

IF[Bug]: "allowed_openai_params" does not work for SDK (thinking param)

THENTrying to use LiteLLM SDK to connect to upstream OpenAI-compatible provider (actually, another LiteLLM proxy) and pass "thinking" parameter for the claude models, adding "allowed_openai_params": ["thinking"].