 Alexander Korolev
|
803f77a43c
refactor: enforce explicit imports via #![deny(clippy::wildcard_imports)]
|
4 ヶ月 前 |
|
Alexander Korolev
|
679c256a60
refactor: resolved_hf_cache_dir returns PathBuf instead of String
|
4 ヶ月 前 |
|
Alexander Korolev
|
c3f680e662
refactor: change ssh_key_path from String to PathBuf
|
4 ヶ月 前 |
|
Alexander Korolev
|
cd8d9d66d5
refactor: replace wildcard imports with explicit imports (docker/mod.rs)
|
4 ヶ月 前 |
|
Alexander Korolev
|
fc645f8d70
feat: add 'power switch --from <src> --to <tgt>' command
|
4 ヶ月 前 |
|
Alexander Korolev
|
17f619fb30
feat: prefix matching for cluster names (--cluster 2 → 2nodes-dac)
|
4 ヶ月 前 |
|
Alexander Korolev
|
68132ce91a
feat: set TERM=xterm-256color on head node for colored logs
|
4 ヶ月 前 |
|
Alexander Korolev
|
c62934d8c3
chore: cleanup
|
4 ヶ月 前 |
|
Alexander Korolev
|
227eb18915
feat: auto-set VLLM_LOGGING_COLOR=1 on head node (override via recipe [env])
|
4 ヶ月 前 |
|
Alexander Korolev
|
338828e55d
feat: force colored vLLM logs via VLLM_LOGGING_COLOR=1
|
4 ヶ月 前 |
|
Alexander Korolev
|
75ffd84ad2
feat(minimax-m3-mxfp8): alias minimax-m3
|
4 ヶ月 前 |
|
Alexander Korolev
|
1dfee6da47
feat: add MiniMax-M3-MXFP8 recipe (vLLM, 4 nodes, TP=4)
|
4 ヶ月 前 |
|
Alexander Korolev
|
2cf821840c
fix(deepseek): comment out shm_size (redundant with --privileged)
|
4 ヶ月 前 |
|
Alexander Korolev
|
7b01e19ec2
fix: HF_HOME -> /root/.cache/huggingface (fixes stale cache)
|
4 ヶ月 前 |
|
Alexander Korolev
|
9241bdc682
fix: don't pass --distributed-executor-backend mp (auto-detected)
|
4 ヶ月 前 |
|
Alexander Korolev
|
6f2c33a82f
fix: only add --shm-size when recipe explicitly sets it
|
4 ヶ月 前 |
|
Alexander Korolev
|
bc41cdf4c9
fix(deepseek): restore --reasoning-config and reasoning_effort
|
4 ヶ月 前 |
|
Alexander Korolev
|
d9c3f845c2
fix(deepseek): align parameters with working system
|
4 ヶ月 前 |
|
Alexander Korolev
|
8df3f532b7
fix(deepseek): restore clean literal strings for JSON configs
|
4 ヶ月 前 |
|
Alexander Korolev
|
b0c139ea80
feat: auto-select smallest cluster matching required_nodes
|
4 ヶ月 前 |
|
Alexander Korolev
|
a9a35440a0
chore: cleanup recipes
|
4 ヶ月 前 |
|
Alexander Korolev
|
f5f4a2ae73
docs(deepseek): use TOML literal strings for JSON configs
|
4 ヶ月 前 |
|
Alexander Korolev
|
d521cfd724
docs(deepseek): use triple-quoted strings for JSON configs
|
4 ヶ月 前 |
|
Alexander Korolev
|
2af1104e7f
feat: launcher auto-adds --host/--port; recipes no longer hardcode
|
4 ヶ月 前 |
|
Alexander Korolev
|
1fa2811301
feat: default recipe port is 30000
|
4 ヶ月 前 |
|
Alexander Korolev
|
7ba894be67
feat: per-recipe shm_size + DeepSeek V4 Flash recipe
|
4 ヶ月 前 |
|
Alexander Korolev
|
7b9f963036
feat: optional cluster-level extra_entities for power commands
|
4 ヶ月 前 |
|
Alexander Korolev
|
7b8651cead
feat: Home Assistant integration for power on/off
|
4 ヶ月 前 |
|
Alexander Korolev
|
9c0e7680a4
feat: set safe NCCL defaults (IGNORE_CPU_AFFINITY, NET, CROSS_NIC, DEBUG)
|
4 ヶ月 前 |
|
Alexander Korolev
|
77a77c0f32
docs: document every environment variable the launcher passes
|
4 ヶ月 前 |
