ExLlamaSharp v1.3.2.1 is out.
Local LLM server for Windows with NVIDIA GPUs — OpenAI-compatible /v1, Blazor admin, and EXL3 inference.
Release notes
Summary - LAN access toggle in Settings/Setup with firewall rule helper (Tray elevated netsh). - Worker hardening: prompt_too_long / kv_cache_full fail-fast, Triton warmup, max_new clamp to remaining context. - Clearer OpenAI timeout messaging; default max tokens / batch context tuning for 12 GB GPUs. ## Install Download ExLlamaSharp-Setup-win-x64.exe, run as Administrator, choose Desktop mode.
Full notes: https://github.com/Kortexio/ExLlamaSharp/releases/tag/v1.3.2.1
Top comments (0)