deploy parry

BADGES.md

@@ -0,0 +1,14 @@
+# 🎖️ Parry — badge evidence, one click each
+
+*(Duplicated from README.md so it can't be missed: this file, the README table,
+the in-game "✦ artifacts" links, and the `about` Gradio API all carry the same
+links.)*
+
+| Badge | Evidence |
+|---|---|
+| 🔌 **Off the Grid** | All inference runs in YOUR browser: Tactician + imagination on WebGPU (WebLLM), the tuned Analyst on llama.cpp/WASM. `GET /healthz` → `"inference": "in-browser"` — pull your Wi-Fi out mid-match. |
+| 🎯 **Well-Tuned** | The published fine-tune **runs inside this app** as the Analyst. [parry-tactician-1.5b-merged](https://huggingface.co/Jainamshahhh/parry-tactician-1.5b-merged) — model card has the recipe, the six-generation eval arc (including the failed gates), the CFG γ-sweep table, and `evals/` with all 18 raw report JSONs. Adapter: [-lora](https://huggingface.co/Jainamshahhh/parry-tactician-1.5b-lora). |
+| 🦙 **Llama Champion** | [parry-tactician-1.5b-gguf](https://huggingface.co/Jainamshahhh/parry-tactician-1.5b-gguf) (Q4_K_M) runs through **llama.cpp's WASM runtime (wllama)** in-browser. Watch for "✦ tuned analyst (llama.cpp)" in the HUD meta line. |
+| 🎨 **Off-Brand** | The custom pixel-dojo canvas, served by `gr.Server` — Gradio's engine at `/`, the `trace_digest` endpoint through Gradio's queue. |
+| 📡 **Sharing is Caring** | [parry-traces](https://huggingface.co/datasets/Jainamshahhh/parry-traces) — decision-by-decision agent traces with full provenance, replayable against the deterministic engine. Export your own: **B → Export trace**. |
+| 📓 **Field Notes** | [The full build report](https://huggingface.co/datasets/Jainamshahhh/parry-field-notes) — six model generations, three honestly-failed pre-registered gates, the engine-physics bug a playtest found, classifier-free guidance for game policies, and the pose-mode saga. |

app.py

@@ -101,6 +101,14 @@ def _about() -> dict:
         "fallback": "none — inference is 100% in-browser by design",
         "tuned_model": "Jainamshahhh/parry-tactician-1.5b-lora (published fine-tune)",
         "thesis": "a sub-100ms reaction loop no network round-trip can serve",
+        "artifacts": {
+            "fine_tune_merged": "https://huggingface.co/Jainamshahhh/parry-tactician-1.5b-merged",
+            "fine_tune_lora": "https://huggingface.co/Jainamshahhh/parry-tactician-1.5b-lora",
+            "gguf_llama_cpp": "https://huggingface.co/Jainamshahhh/parry-tactician-1.5b-gguf",
+            "field_notes": "https://huggingface.co/datasets/Jainamshahhh/parry-field-notes",
+            "agent_traces": "https://huggingface.co/datasets/Jainamshahhh/parry-traces",
+            "badge_evidence": "see README.md / BADGES.md in this Space repo",
+        },
     }
 
 

static/index.html

@@ -13,7 +13,7 @@
       canvas { display: block; width: 100%; height: 100%; image-rendering: pixelated; }
       #hud { position: absolute; inset: 0; pointer-events: none; }
     </style>
-    <script type="module" crossorigin src="./assets/main-qSw1hafF.js"></script>
+    <script type="module" crossorigin src="./assets/main-gooejseS.js"></script>
     <link rel="modulepreload" crossorigin href="./assets/action_set-C6mnKwO1.js">
     <link rel="modulepreload" crossorigin href="./assets/engine-DIUq88up.js">
     <link rel="modulepreload" crossorigin href="./assets/index-K8o-xskJ.js">