 Alexander Korolev
|
c62934d8c3
chore: cleanup
|
vor 4 Monaten |
|
Alexander Korolev
|
227eb18915
feat: auto-set VLLM_LOGGING_COLOR=1 on head node (override via recipe [env])
|
vor 4 Monaten |
|
Alexander Korolev
|
338828e55d
feat: force colored vLLM logs via VLLM_LOGGING_COLOR=1
|
vor 4 Monaten |
|
Alexander Korolev
|
75ffd84ad2
feat(minimax-m3-mxfp8): alias minimax-m3
|
vor 4 Monaten |
|
Alexander Korolev
|
1dfee6da47
feat: add MiniMax-M3-MXFP8 recipe (vLLM, 4 nodes, TP=4)
|
vor 4 Monaten |
|
Alexander Korolev
|
2cf821840c
fix(deepseek): comment out shm_size (redundant with --privileged)
|
vor 4 Monaten |
|
Alexander Korolev
|
7b01e19ec2
fix: HF_HOME -> /root/.cache/huggingface (fixes stale cache)
|
vor 4 Monaten |
|
Alexander Korolev
|
9241bdc682
fix: don't pass --distributed-executor-backend mp (auto-detected)
|
vor 4 Monaten |
|
Alexander Korolev
|
6f2c33a82f
fix: only add --shm-size when recipe explicitly sets it
|
vor 4 Monaten |
|
Alexander Korolev
|
bc41cdf4c9
fix(deepseek): restore --reasoning-config and reasoning_effort
|
vor 4 Monaten |
|
Alexander Korolev
|
d9c3f845c2
fix(deepseek): align parameters with working system
|
vor 4 Monaten |
|
Alexander Korolev
|
8df3f532b7
fix(deepseek): restore clean literal strings for JSON configs
|
vor 4 Monaten |
|
Alexander Korolev
|
b0c139ea80
feat: auto-select smallest cluster matching required_nodes
|
vor 4 Monaten |
|
Alexander Korolev
|
a9a35440a0
chore: cleanup recipes
|
vor 4 Monaten |
|
Alexander Korolev
|
f5f4a2ae73
docs(deepseek): use TOML literal strings for JSON configs
|
vor 4 Monaten |
|
Alexander Korolev
|
d521cfd724
docs(deepseek): use triple-quoted strings for JSON configs
|
vor 4 Monaten |
|
Alexander Korolev
|
2af1104e7f
feat: launcher auto-adds --host/--port; recipes no longer hardcode
|
vor 4 Monaten |
|
Alexander Korolev
|
1fa2811301
feat: default recipe port is 30000
|
vor 4 Monaten |
|
Alexander Korolev
|
7ba894be67
feat: per-recipe shm_size + DeepSeek V4 Flash recipe
|
vor 4 Monaten |
|
Alexander Korolev
|
7b9f963036
feat: optional cluster-level extra_entities for power commands
|
vor 4 Monaten |
|
Alexander Korolev
|
7b8651cead
feat: Home Assistant integration for power on/off
|
vor 4 Monaten |
|
Alexander Korolev
|
9c0e7680a4
feat: set safe NCCL defaults (IGNORE_CPU_AFFINITY, NET, CROSS_NIC, DEBUG)
|
vor 4 Monaten |
|
Alexander Korolev
|
77a77c0f32
docs: document every environment variable the launcher passes
|
vor 4 Monaten |
|
Alexander Korolev
|
1b02f53b07
feat: set GLOO_SOCKET_IFNAME and TP_SOCKET_IFNAME for vLLM
|
vor 4 Monaten |
|
Alexander Korolev
|
071837e830
feat: fallback to ~/.config/sm121-launcher/ for config + recipes
|
vor 4 Monaten |
|
Alexander Korolev
|
ce40c23b7a
feat: section aliases for exec subcommand
|
vor 4 Monaten |
|
Alexander Korolev
|
bfe94fa39e
feat: pre-serve hooks + exec subcommand for cluster mass-execution
|
vor 4 Monaten |
|
Alexander Korolev
|
5097d55c77
fix: redirect stderr in `cargo run logs --follow` docker pipeline
|
vor 4 Monaten |
|
Alexander Korolev
|
65cbf60fce
build: enforce quality gates via scripts/check.sh
|
vor 4 Monaten |
|
Alexander Korolev
|
df5e3a701e
feat: add 'init-host-keys' command to auto-write SSH host_keys
|
vor 4 Monaten |
