Open Issues Need Help
View All on GitHub [Qwen3-Omni Perf] Code2Wav: CUDA-graph shape hit-rate telemetry about 4 hours ago
good first issue
SGLang Omni: High-Performance Multi-Stage Pipeline Framework for Omni Models
Python
[Qwen3-Omni Perf] Predictor attention: use SDPA `enable_gqa` instead of materialized KV expansion about 4 hours ago
good first issue
SGLang Omni: High-Performance Multi-Stage Pipeline Framework for Omni Models
Python
[Qwen3-Omni Perf] Thinker prefill: batch the multimodal embedding merge, remove per-request host syncs about 5 hours ago
good first issue
SGLang Omni: High-Performance Multi-Stage Pipeline Framework for Omni Models
Python
[Qwen3-Omni Perf] Thinker: vectorize MRoPE position computation; skip multimodal positions on speech stages about 5 hours ago
good first issue
SGLang Omni: High-Performance Multi-Stage Pipeline Framework for Omni Models
Python
[Qwen3-Omni Perf] Code2Wav: wire configurable initial codec chunk size into the Qwen3-Omni vocoder scheduler about 5 hours ago
good first issue
SGLang Omni: High-Performance Multi-Stage Pipeline Framework for Omni Models
Python
good first issue
SGLang Omni: High-Performance Multi-Stage Pipeline Framework for Omni Models
Python
[Good First Issue] Optimize Qwen3-ASR for faster TTS WER evaluation about 1 month ago
good first issue
SGLang Omni: High-Performance Multi-Stage Pipeline Framework for Omni Models
Python
[Bug] OpenAI /v1/chat/completions text-only streaming emits a single chunk (no per-token TPOT) about 1 month ago
good first issue
SGLang Omni: High-Performance Multi-Stage Pipeline Framework for Omni Models
Python