sirius0xdev
|
2f163fab36
|
Merge pull request #46 from sirius0xdev/feature/maximize-l4-context
Maximize L4 vLLM context window
|
2026-04-20 12:29:03 -04:00 |
|
Sirius Claw
|
fb951600d7
|
chore: maximize vllm max_model_len on L4 to 32768
|
2026-04-20 16:14:27 +00:00 |
|
Sirius Claw
|
687d61f626
|
fix: pin vLLM to v0.9.1 and use pythonic tool call parser
|
2026-04-20 07:23:43 +00:00 |
|
Sirius Claw
|
baa88975b5
|
fix: revert vLLM image to latest to support qwen3_coder tool parser
|
2026-04-20 07:16:42 +00:00 |
|
Sirius Claw
|
35dc7f01ac
|
fix: bump pinned vLLM version to v0.7.3 to support tool calling arguments
|
2026-04-20 06:58:13 +00:00 |
|
Sirius Claw
|
43eb51d5b7
|
feat: optimize cost via L4 spot instances and vLLM performance tuning
|
2026-04-20 05:38:36 +00:00 |
|
Sirius Claw
|
7d34b3f649
|
fix: match vllm-l4 deployment nodeSelector and tolerations to nodepool labels/taints
|
2026-04-20 05:11:55 +00:00 |
|
Sirius Claw
|
8041f71063
|
feat: Switch L4 deployment to Qwen2.5-Coder-7B-Instruct-heretic
|
2026-04-20 02:32:01 +00:00 |
|
Sirius Claw
|
7a0d33f042
|
feat: Add L4 vLLM deployment and KEDA scale-to-zero for A100
|
2026-04-20 02:20:08 +00:00 |
|