fffiloni commited on
Commit
58f8f62
·
verified ·
1 Parent(s): 99a2759

Upload 3 files

Browse files
Files changed (2) hide show
  1. docs/ARCHITECTURE.md +10 -0
  2. docs/NEXT_STEPS.md +10 -0
docs/ARCHITECTURE.md CHANGED
@@ -45,3 +45,13 @@ The target Space is private by default and must be created in the signed-in user
45
  Before adding Pi, model-card analysis, or ZeroGPU, we need to prove that the system can safely create a target Space and validate it through the live API.
46
 
47
  That live API validation gate will later remain the core success criterion.
 
 
 
 
 
 
 
 
 
 
 
45
  Before adding Pi, model-card analysis, or ZeroGPU, we need to prove that the system can safely create a target Space and validate it through the live API.
46
 
47
  That live API validation gate will later remain the core success criterion.
48
+
49
+
50
+ ## Phase 6
51
+
52
+ Adds a no-build runtime recommender Job that analyzes model metadata and writes `runtime_recommendation.json` to the Bucket.
53
+
54
+
55
+ ## Phase 7 — LongCat article reproduction
56
+
57
+ Phase 7 attempts an article-style LongCat Space build: private target Space, Pi-guided app adaptation, ZeroGPU first, fixed GPU fallback when explicitly enabled, and live `/health` API validation. Full video generation remains a manual-review step until model-specific runtime validation is complete.
docs/NEXT_STEPS.md CHANGED
@@ -31,3 +31,13 @@ Keep these out of scope for Phase 5:
31
  - ComfyUI
32
  - video generation
33
  - public publication automation
 
 
 
 
 
 
 
 
 
 
 
31
  - ComfyUI
32
  - video generation
33
  - public publication automation
34
+
35
+
36
+ ## Phase 6
37
+
38
+ Adds a no-build runtime recommender Job that analyzes model metadata and writes `runtime_recommendation.json` to the Bucket.
39
+
40
+
41
+ ## Phase 7 — LongCat article reproduction
42
+
43
+ Phase 7 attempts an article-style LongCat Space build: private target Space, Pi-guided app adaptation, ZeroGPU first, fixed GPU fallback when explicitly enabled, and live `/health` API validation. Full video generation remains a manual-review step until model-specific runtime validation is complete.