Add model card

README.md

@@ -42,8 +42,8 @@ official Q4_K_M GGUF and reached **35.62 tokens/second** during the complete
 > [!IMPORTANT]
 > This GGUF uses ROCmFP4 tensor types and Laguna architecture support. It is
 > built for the Laguna-enabled
-> [Ciru ROCmFPX Runtime V3](https://github.com/ciru-ai/ROCmFPX/tree/e4db35e027d0a9fa60e3324e4cc12f63ba7e0a00)
-> at commit `e4db35e027d0a9fa60e3324e4cc12f63ba7e0a00`. It does not load in
+> [Ciru ROCmFPX Runtime V3](https://github.com/ciru-ai/ROCmFPX/tree/54f5fe06c74350fb8b6aec21d8749071bc195bdb)
+> at commit `54f5fe06c74350fb8b6aec21d8749071bc195bdb`. It does not load in
 > stock upstream llama.cpp.
 >
 > Runtime V3 fixes Laguna tool-call grammar/parser agreement and removes a
@@ -80,6 +80,10 @@ Runtime V3 adds parser and grammar fixes on top of the V2 Vulkan runtime:
 - aligns PEG JSON-array comma whitespace handling with the generated grammar;
 - adds a Laguna/Pi regression for an `edit` call containing an
   `array<object>` argument and source-code strings;
+- keeps server checkpoints host-backed, preventing the on-device checkpoint
+  destruction fault seen when a long hybrid/SWA context is invalidated;
+- explicitly disables context checkpoints in the validated Laguna launcher
+  while leaving the normal KV and prompt caches available;
 - retains the V2 RADV DeviceLost safeguards and validated 128K defaults.
 
 Focused V3 validation on Ryzen AI Max+ 395 / Radeon 8060S with Mesa RADV
@@ -90,6 +94,13 @@ smokes returned `Paris.`, preserved the nested `edit` tool call, completed an
 the slot after cancellation at 82% of a 15,000-token request. A follow-up
 request returned `Paris.` with no `VK_ERROR_DEVICE_LOST`.
 
+The checkpoint repair was exercised separately with checkpoints explicitly
+re-enabled and prompt-cache RAM left at 8192 MiB. A 120,045-token request
+created 16 hybrid/SWA checkpoints; an unrelated follow-up forced `pos_next=0`
+and erased all 16, then returned `OK.`. The server remained healthy and shut
+down cleanly without a core dump. Checkpoints nevertheless remain disabled in
+the public profile until this gate is repeated across multi-turn workloads.
+
 ### V2 Vulkan stability baseline
 
 The first runtime release could lose the Vulkan device during a very deep
@@ -209,9 +220,9 @@ Clone and pin Runtime V3 exactly:
 git clone --branch agent/laguna-s21-runtime-v3 --depth 1 \
   https://github.com/ciru-ai/ROCmFPX.git
 cd ROCmFPX
-git checkout --detach e4db35e027d0a9fa60e3324e4cc12f63ba7e0a00
+git checkout --detach 54f5fe06c74350fb8b6aec21d8749071bc195bdb
 test "$(git rev-parse HEAD)" = \
-  "e4db35e027d0a9fa60e3324e4cc12f63ba7e0a00"
+  "54f5fe06c74350fb8b6aec21d8749071bc195bdb"
 ```
 
 The distro-aware helper prints the native package command before making any
@@ -285,7 +296,7 @@ build-laguna-strix-vulkan/bin/test-llama-archs
 ```
 
 The
-[complete Linux and V3 guide](https://github.com/ciru-ai/ROCmFPX/blob/e4db35e027d0a9fa60e3324e4cc12f63ba7e0a00/docs/recipes/laguna-s21-chadrock-rocmfp4-strixkvspine-v4.md)
+[complete Linux and V3 guide](https://github.com/ciru-ai/ROCmFPX/blob/54f5fe06c74350fb8b6aec21d8749071bc195bdb/docs/recipes/laguna-s21-chadrock-rocmfp4-strixkvspine-v4.md)
 contains the Fedora, Arch, and NixOS commands.
 
 ### Start Laguna with the validated 128K V3 profile
@@ -301,7 +312,7 @@ scripts/run-laguna-vulkan-supervised.sh \
 The launcher applies the measured single-slot Strix Halo configuration:
 Vulkan0, full offload, row split, Flash Attention, 131,072 context, F16/F16
 KV, batch 2048, ubatch 512, node cap 10, FA dispatch width 4, 16 threads,
-thinking off, and this sampler:
+thinking off, context checkpoints disabled, and this sampler:
 
 ```json
 {
@@ -327,6 +338,13 @@ The direct runner uses the same V3 safe defaults without supervision:
 scripts/run-laguna-s21-rocmfp4-v4.sh /path/to/model.gguf
 ```
 
+Context checkpoints are separate from the normal KV cache. The launcher now
+passes `--ctx-checkpoints 0` because the hybrid/SWA checkpoint path has not yet
+completed the full repeated 128K multi-turn and cache-replay qualification.
+The runtime also moves checkpoint payloads back to host memory, matching the
+safer upstream design. Advanced diagnostic runs can opt in with
+`CTX_CHECKPOINTS=N`; this is not part of the validated public profile yet.
+
 The model's 256K capacity remains available only as an explicit experimental
 lane:
 
@@ -339,7 +357,7 @@ That command prints a warning because 256K has not yet passed the
 full-depth prefill, multi-turn, and cache-replay gates.
 
 The complete production recipe is preserved in the
-[ROCmFPX Laguna Runtime V3 guide](https://github.com/ciru-ai/ROCmFPX/blob/e4db35e027d0a9fa60e3324e4cc12f63ba7e0a00/docs/recipes/laguna-s21-chadrock-rocmfp4-strixkvspine-v4.md).
+[ROCmFPX Laguna Runtime V3 guide](https://github.com/ciru-ai/ROCmFPX/blob/54f5fe06c74350fb8b6aec21d8749071bc195bdb/docs/recipes/laguna-s21-chadrock-rocmfp4-strixkvspine-v4.md).
 
 ## Example request