Commit Graph
34 Commits
Author SHA1 Message Date
Mikei386 302b08e051 Add CPU embeddings for OpenClaw memory 2026-09-21 08:31:45 +02:00
Mikei386 b48fcc63d3 Tune prefill batching across Qwen profiles 2026-09-15 16:57:53 +02:00
Mikei386 78055996f5 Increase medium prefill micro-batch 2026-09-15 16:36:48 +02:00
Mikei386 fce9900389 Synchronize repository with Athena deployment 2026-09-13 20:01:36 +02:00
Mikei386 cc416150a8 Replace XTTS with Qwen3-TTS 2026-09-05 13:30:14 +02:00
Mikei386 f553108912 Add Qwen GSQ-RCO Beta 1 profile 2026-09-04 14:36:50 +02:00
Mikei386 fb0cb40bed Increase llama prompt cache to 32 GiB 2026-09-03 10:03:43 +02:00
Mikei386 e30f4f024c Fix Athena recovery and installation consistency 2026-09-01 10:19:13 +02:00
Mikei386 c031fd2c55 Remove obsolete experimental profile 2026-09-01 09:04:45 +02:00
Mikei386 91fbb276bd Remove obsolete review model service 2026-09-01 08:54:33 +02:00
Mikei386 211ede9fc5 Update llama runtime and isolate background reviews 2026-08-31 21:11:17 +02:00
Mikei386 7e36f1dbb7 Fix Thinking Off handling 2026-08-31 08:15:27 +02:00
Mikei386 0fa9a34886 Increase llama prompt cache to 24 GiB 2026-08-30 22:33:37 +02:00
Mikei386 a41d6dc28c Enable benchmarked two-slot medium profile 2026-08-30 20:54:22 +02:00
Mikei386 db3719af7a Default reasoning effort to medium 2026-08-30 20:08:32 +02:00
Mikei386 2661553e20 Tune prefill batching across Qwen profiles 2026-08-30 19:50:40 +02:00
Mikei386 edb8b2868f Tune Qwen medium prefill batching 2026-08-30 18:39:17 +02:00
Mikei386 6af25cf30c Add local FLUX image editing 2026-08-30 10:07:40 +02:00
Mikei386 c6517ee137 Simplify Athena runtime and document current architecture 2026-08-30 08:45:52 +02:00
Mikei386 896c12193e Replace compression sidecar with Hermes micro-compaction 2026-08-26 06:18:51 +02:00
Mikei386 aaadc57c07 Harden and demote experimental Hermes compressor 2026-08-26 01:40:38 +02:00
Mikei386 43d9903d13 Add dedicated Hermes compression model 2026-08-26 00:45:32 +02:00
Mikei386 4ac0c479c4 fix(openwebui): synthesize after six tool calls 2026-08-24 08:43:14 +02:00
Mikei386 50c37b09bb fix(openwebui): bound calls and force prose synthesis 2026-08-24 08:33:44 +02:00
Mikei386 4d4eb0b8d8 fix(openwebui): pin migrated database-compatible base 2026-08-24 08:14:27 +02:00
Mikei386 f90fc93f9c Add XTTS primary voice with Piper fallback 2026-08-23 12:22:59 +02:00
Mikei386 2ef894160a Add final Qwen3.8 runtime and uncensored profile 2026-08-23 00:10:53 +02:00
Mikei386 8267a85a96 Offload vision projector to RTX 3060 2026-08-22 17:19:56 +02:00
Mikei386 41b9c17f0f Define four-profile production matrix with Medium default 2026-08-22 11:30:38 +02:00
Mikei386 977f8f5c76 Validate Pure quant for 256K Ultra profile 2026-08-22 11:01:13 +02:00
Mikei386 f52cf14069 Add tested 256K Ultra profile 2026-08-22 10:44:14 +02:00
Mikei386 a5cf11582a Add reproducible Piper TTS service 2026-08-21 17:02:32 +02:00
Mikei386 42e2aad3a6 Use embedded MTP tensor for IQ4-MIX profiles 2026-08-21 13:18:45 +02:00
Mikei386 e83c0e2c70 Add reproducible Docker and WireGuard host bootstrap 2026-08-20 21:23:16 +02:00