Repository Issue Activity (beta)

#909 Include OTel trace_id/span_id in JSON logs for log-trace correlation

#922 Documentation proposal: Kubernetes deployment with a community Helm chart

#889 fix: unbounded buffer allocation in paddedbatch.from_pb

#900 Publish arm64 images with release numbers (e.g. cpu-arm64-1.9)

#879 te_request_count (method=single) is incremented twice for single requests

#424 thread 'tokio-runtime-worker' panicked at /usr/src/backends/src/lib.rs:176:14

#870 cuda-entrypoint.sh awk pattern fails on NVIDIA driver 6xx (CUDA UMD Version: header rename) → CUDA_ERROR_NO_DEVICE → CPU fallback

#815 bfloat16 support

#914 Non-flash F32 path fails with CUDA_ERROR_INVALID_VALUE on sm_120 with candle 0.9.2 (#809 branch)

#911 Feature request: OTLP over HTTP (http-proto) exporter for OpenTelemetry traces

#333 Images Embeddings (ex. CLIP model)

#902 target-cpu=native causes SIGILL when running AMD-built binaries on Intel CPUs

#141 Support for BAAI/bge-m3 model

#895 Header content-range missing during Artifactory proxy download

#888 Support --root-path CLI Argument

#691 Any plan to support qwen3 reranker model in TEI

#643 Feature Request: Add Support for Qwen3-Reranker Model

#763 Qwen3-Reranker-0.6B start error

#881 Health path isn't ignored for Opentelemetry.

#819 Decreasing warmup for use on spot instances

#882 Qwen3/Gemma3 candle skip attention masks for equal-length batches

#880 te_queue_size metric does not increase when requests are appended to the queue

#874 Incorrect retrieval results and inflated similarity scores for sdadas/stella-pl-retrieval-8k in TEI

#861 The PPLX 0.6B model fails to fit within the memory capacity of an A100, resulting in out-of-memory errors.

#872 Vera rubin support