12 Commits
Author SHA1 Message Date
hermenix ce92096042 llama-cpp: llama-server with DFlash2 speculative decoding for gfx1151
v0.8.0 is the first upstream release with DFlash2 drafters; nixpkgs has 0.6.0 (unstable) and an older tree (stable), so the engine is built here. HIP build for Strix Halo. The NixOS module exposes llama-server behind skyg.nixos.server.services.llama-cpp; fwdesk wires it to Qwen3.8-27B + its DFlash2 drafter, disabled until the weights are on disk.
2026-11-11 16:01:32 +00:00
ari e3f01e3c35 cleaning up old bots 2026-11-05 15:21:01 -05:00
ari a60b842ee2 more nbot stuff 2026-11-03 19:21:50 -05:00
ari dd4965f030 bazel 2026-11-01 14:26:16 -05:00
ari 507d603ef0 clean up of ds4 (using my fork of it instead) + update scripts!!! 2026-09-25 10:25:13 -04:00
ari 4cc2123eff buzz + update ds4 2026-08-05 16:42:57 -04:00
ari 2e940cc2c8 using container systemd services 2026-07-08 22:15:13 -04:00
ari 3cb0b970bf ds4 server 2026-07-01 18:06:52 -04:00
ari 50ef9c7238 signal 2026-06-26 16:59:18 -04:00
ari 92df288796 comfyui 2026-04-15 12:57:46 -04:00
ari c6548fb88c dockge 2024-11-09 11:20:23 -05:00
ari 52ae9d193b refactor services 2024-11-09 11:12:19 -05:00