valknar

valknar pushed to main at valknar/supervisor-ui

2025-11-23 18:24:03 +01:00

e0cfd371c0 feat: initial commit - Supervisor UI with Next.js 16 and Tailwind CSS 4

valknar pushed to main at valknar/runpod

2025-11-23 18:16:38 +01:00

5944767d3f fix: update CivitAI model version IDs to latest versions

valknar pushed to main at valknar/bin

2025-11-23 17:59:59 +01:00

a1548ed490 fix: correct typo in filename (hugginface -> huggingface)

valknar pushed to main at valknar/bin

2025-11-23 17:58:52 +01:00

6b2068e803 feat: add CivitAI NSFW model downloader script

valknar pushed to main at valknar/runpod

2025-11-23 17:58:32 +01:00

e29f77c90b feat: add dedicated CivitAI NSFW model downloader

valknar created repository valknar/supervisor-ui

2025-11-23 17:50:59 +01:00

valknar pushed to main at valknar/runpod

2025-11-23 16:27:02 +01:00

76cf5b5e31 docs: update CLAUDE.md to reflect direct vLLM architecture

valknar pushed to main at valknar/runpod

2025-11-23 16:24:44 +01:00

479201d338 chore: remove orchestrator - replaced with dedicated vLLM servers

valknar pushed to main at valknar/docker-compose

2025-11-23 16:22:57 +01:00

a80c6b931b fix: update compose.yaml to use new GPU_VLLM URLs

valknar pushed to main at valknar/docker-compose

2025-11-23 16:17:30 +01:00

64c02228d8 fix: use EMPTY api_key for vLLM servers

valknar pushed to main at valknar/docker-compose

2025-11-23 16:16:39 +01:00

55d9bef18a fix: remove api_key from vLLM config to fix authentication error

valknar pushed to main at valknar/docker-compose

2025-11-23 16:10:25 +01:00

7fc945e179 fix: update LiteLLM config for direct vLLM server access

valknar pushed to main at valknar/runpod

2025-11-23 16:00:06 +01:00

1ad99cdb53 refactor: replace orchestrator with dedicated vLLM servers for Qwen and Llama

valknar pushed to main at valknar/runpod

2025-11-23 15:43:40 +01:00

cc0f55df38 fix: reduce max_model_len to 20000 to fit in 24GB VRAM

valknar pushed to main at valknar/runpod

2025-11-23 15:38:21 +01:00

5cfd03f1ef fix: improve streaming with proper delta format and increase max_model_len to 32768

valknar pushed to main at valknar/runpod

2025-11-23 15:21:54 +01:00

3f812704a2 fix: use venv python for vLLM service startup

valknar pushed to main at valknar/runpod

2025-11-23 15:10:08 +01:00

fdd724298a fix: increase max_tokens limit from 4096 to 32768 for LLMX CLI support

valknar pushed to main at valknar/docker-compose

2025-11-23 14:36:39 +01:00

94ab4ae6dd feat: enable system message support for qwen-2.5-7b

valknar pushed to main at valknar/runpod

2025-11-23 13:45:05 +01:00

a8c2ee1b90 fix: make model name and port configurable via environment variables

valknar pushed to main at valknar/runpod

2025-11-23 13:33:49 +01:00

16112e50f6 fix: relax dependency version constraints for vllm compatibility