Update ISO installer tiers and backend deployment
This commit is contained in:
1 parent
79784a6743
commit
56668f7bdc
38 files changed
+7206
-175
No files matched your search
@@ -0,0 +1,7 @@
|
||||
---
|
||||
# vLLM role defaults — overridden per-tier in the tier playbook vars block
|
||||
vllm_model: "meta-llama/Meta-Llama-3.1-8B-Instruct"
|
||||
vllm_tensor_parallel: 1
|
||||
vllm_gpu_memory_util: "0.70"
|
||||
vllm_max_model_len: 8192
|
||||
vllm_quantization: "" # blank = full precision; set to "awq" for 4-bit
|
||||
Reference in new issue
Block a user