-
Notifications
You must be signed in to change notification settings - Fork 3
All issues
Issue creation is restricted in this repository
Issues
is:issue state:open
is:issue state:open
Search results
- Status: Open.#88 In LucknowAI/loxi_llm;
- Status: Open.#86 In LucknowAI/loxi_llm;
- Status: Open.#84 In LucknowAI/loxi_llm;
- Status: Open.#80 In LucknowAI/loxi_llm;
[Perf] Redundant candidate-buffer refill in grammar-resample path of sampleNextToken
agentAgent harnessAgent harnessllama_enginellama_engine pluginllama_engine pluginnativeNative / JNI workNative / JNI workperformancePerformancePerformanceStatus: Open.#75 In LucknowAI/loxi_llm;[Chore] Streaming state guard-then-copy pattern duplicated 5x in chat_notifier.dart
chatChat featureChat featureStatus: Open.#74 In LucknowAI/loxi_llm;[Chore] Generation timeout fallback duplicated between send() and _runAgentTurn with different values
agentAgent harnessAgent harnesschatChat featureChat featureStatus: Open.#72 In LucknowAI/loxi_llm;[Perf] Base model and mmproj companion download sequentially instead of concurrently
modelsModel catalogModel catalogmultimodalMultimodal vision supportMultimodal vision supportperformancePerformancePerformanceStatus: Open.#71 In LucknowAI/loxi_llm;[Bug] RAM check silently undercounts a multimodal model's footprint if mmprojSizeBytes is missing
bugSomething isn't workingSomething isn't workingmodelsModel catalogModel catalogmultimodalMultimodal vision supportMultimodal vision supportStatus: Open.#70 In LucknowAI/loxi_llm;[Bug] Unawaited backend.stop() on timeout can race the next agent-loop iteration
agentAgent harnessAgent harnessbugSomething isn't workingSomething isn't workingchatChat featureChat featureStatus: Open.#69 In LucknowAI/loxi_llm;[Multimodal] Stop/timeout does not interrupt in-flight vision encoding
llama_enginellama_engine pluginllama_engine pluginmultimodalMultimodal vision supportMultimodal vision supportnativeNative / JNI workNative / JNI workperformancePerformancePerformanceStatus: Open.#58 In LucknowAI/loxi_llm;[Multimodal] Speed up vision first-token latency (CPU-only encode taking minutes)
llama_enginellama_engine pluginllama_engine pluginmultimodalMultimodal vision supportMultimodal vision supportnativeNative / JNI workNative / JNI workperformancePerformancePerformanceStatus: Open.#57 In LucknowAI/loxi_llm;