compile cache model card update

compile_cache_model_card.md

@@ -1,6 +1,6 @@
 # Compile Cache — MI300X
 
-Generated: 2026-05-18T05:28:54Z
+Generated: 2026-05-21T01:16:23Z
 
 ## Purpose
 Records the exact configuration under which the compile cache (inductor + Triton) was built.
@@ -18,7 +18,7 @@ against these values to pinpoint the cache-key mismatch.
 | Torch   | 2.10.0+rocm7.1 |
 | HIP / ROCm | 7.1.25424 |
 | Triton  | 3.6.0 |
-| ck4inductor | 0.0.11 |
+| ck4inductor | <function __version__ at 0x7f81d478afc0> |
 | Python  | 3.11.15 |
 
 ## Compile Configuration
@@ -26,7 +26,7 @@ against these values to pinpoint the cache-key mismatch.
 |---------|-------|
 | Compile backend | inductor |
 | Compile mode | max-autotune-no-cudagraphs |
-| GEMM backends | CK,ATEN |
+| GEMM backends | ATEN |
 | TORCHINDUCTOR\_CK\_DIR | `/home/hotaisle/.local/lib/python3.11/site-packages/ck4inductor` |
 | TORCHINDUCTOR\_CACHE\_DIR | `/workspace/inductor_cache` |
 | TRITON\_CACHE\_DIR | `/workspace/triton_cache` |
@@ -42,19 +42,19 @@ against these values to pinpoint the cache-key mismatch.
 ## Cache Statistics
 | Cache | Files |
 |-------|-------|
-| inductor\_cache | 949 |
-| triton\_cache | 37207 |
-| tar.gz size | 1851.6 MB |
+| inductor\_cache | 372 |
+| triton\_cache | 2295 |
+| tar.gz size | 168.5 MB |
 
 ## Autotune Cache Key
 The inductor autotune winner cache key is a **hash of the full candidate set**
 including which backends are enabled. Adding/removing from
-`TORCHINDUCTOR_MAX_AUTOTUNE_GEMM_BACKENDS` (currently `CK,ATEN`) will
+`TORCHINDUCTOR_MAX_AUTOTUNE_GEMM_BACKENDS` (currently `ATEN`) will
 invalidate all cached winners and force a full AUTOTUNE re-run.
 
 ## Recompilation Trigger Checklist
 If AUTOTUNE blocks appear on a pod that should have cache hits, verify:
-1. `TORCHINDUCTOR_MAX_AUTOTUNE_GEMM_BACKENDS` == `CK,ATEN`
+1. `TORCHINDUCTOR_MAX_AUTOTUNE_GEMM_BACKENDS` == `ATEN`
 2. `TORCHINDUCTOR_CK_DIR` points to valid CK root: `/home/hotaisle/.local/lib/python3.11/site-packages/ck4inductor`
 3. `library/src` symlink inside ck4inductor package exists (needed for gemm_universal kernel enumeration)
 4. `TORCHINDUCTOR_CACHE_DIR` == `/workspace/inductor_cache` (with extracted cache files present)