Reachy DM landing page: write-up + tags + video placeholder

README.md

@@ -1,13 +1,49 @@
 ---
-title: Reachy Dm
-emoji: 🏢
-colorFrom: green
-colorTo: yellow
+title: Reachy DM — Thousand Token Wood
+emoji: 🎲
+colorFrom: purple
+colorTo: indigo
 sdk: gradio
-sdk_version: 6.18.0
-python_version: '3.13'
 app_file: app.py
-pinned: false
+pinned: true
+short_description: Whimsical all-Qwen voice DM, embodied by Reachy Mini
+tags:
+  - track:wood
+  - sponsor:modal
+  - achievement:bestdemo
+  - achievement:bestagent
+  - achievement:offbrand
+  - agent
+  - voice
+  - mcp
+  - tabletop-rpg
+  - qwen
+  - reachy_mini
 ---
 
-Check out the configuration reference at https://huggingface.co/docs/hub/spaces-config-reference
+# 🎲 Reachy DM — Thousand Token Wood
+
+A whimsical AI tabletop **dungeon master you play out loud**, embodied by a **Reachy Mini** robot.
+It narrates scenes, voices NPCs in distinct designed voices, rolls dice, reads your physical
+character sheet through a camera, sets the mood with smart lights, and remembers your party —
+an **all-Qwen, open-weights** stack with every model under the 32B cap.
+
+The full embodied experience (robot + character voices + dice/sheet widgets + story-reactive
+LIFX lighting) is a **local hardware rig**, so this Space is the showcase: the **demo video** and
+a write-up of how it was built. See the page for details.
+
+- **DM brain:** Qwen3-30B-A3B-Instruct-2507 (FP8) on **Modal** (H100)
+- **Character voices:** Qwen3-TTS-12Hz-1.7B on **Modal** (L4)
+- **Vision ("read the table"):** Qwen3-VL-8B on **Modal** (L40S)
+- **Tools via MCP:** dice / character sheet / player choice (rendered as Pip-Boy widgets),
+  per-character voices, durable memory, camera, robot motion, Home Assistant lighting.
+
+**Track:** Thousand Token Wood (whimsical). **Code:**
+[github.com/olaservo/reachy_mini_conversation_app](https://github.com/olaservo/reachy_mini_conversation_app)
+(branch `cascade-integration`).
+
+## Disclaimer
+A personal, non-commercial creative project for the Build Small hackathon, running *Fallout: The
+Roleplaying Game* (Modiphius 2d20) from a copy I own. **Not affiliated with, endorsed by, or
+sponsored by Bethesda Softworks or Modiphius Entertainment.** All rights to those properties
+belong to their respective owners.

app.py

@@ -0,0 +1,96 @@
+"""Reachy DM — Build Small hackathon submission landing page.
+
+A whimsical all-Qwen voice dungeon master you play out loud, embodied by a Reachy Mini.
+The full experience (robot + character voices + dice/sheet widgets + story-reactive lights)
+is a local hardware rig — this Space is the showcase: the demo video + how it was built.
+"""
+
+import gradio as gr
+
+# --- Fill these in after recording / posting ---
+DEMO_VIDEO_EMBED_URL = ""   # e.g. "https://www.youtube.com/embed/XXXXXXXXXXX"
+SOCIAL_POST_URL = ""        # link to the social post
+GITHUB_URL = "https://github.com/olaservo/reachy_mini_conversation_app"
+
+PIPBOY_CSS = """
+.gradio-container { background: radial-gradient(circle at 50% 0%, #0b2a16 0%, #05140a 80%) !important; }
+#pip * { color: #7CFC8A !important; font-family: 'Courier New', monospace !important; }
+#pip h1 { text-shadow: 0 0 10px #2f8; letter-spacing: 1px; }
+#pip a { color: #b6ff8a !important; text-decoration: underline; }
+#pip table { border-color: #2f8a4a !important; }
+#pip code { background: rgba(47,138,74,0.12) !important; border: 1px solid #2f8a4a; padding: 1px 4px; }
+#pip .vault { border: 1px solid #2f8a4a; border-radius: 10px; padding: 14px 18px; background: rgba(47,138,74,0.06); }
+"""
+
+VIDEO_BLOCK = (
+    f'<div class="vault"><iframe width="100%" height="420" src="{DEMO_VIDEO_EMBED_URL}" '
+    'frameborder="0" allow="accelerometer; autoplay; clipboard-write; encrypted-media; '
+    'gyroscope; picture-in-picture" allowfullscreen></iframe></div>'
+    if DEMO_VIDEO_EMBED_URL
+    else '<div class="vault">▶ <b>Demo video coming soon.</b> Check back shortly.</div>'
+)
+
+SOCIAL_LINE = f"**Social post:** [link]({SOCIAL_POST_URL})  ·  " if SOCIAL_POST_URL else ""
+
+WRITEUP = f"""
+# 🎲 Reachy DM — a whimsical AI tabletop dungeon master you play *out loud*
+
+**Track: Thousand Token Wood (whimsical).** Reachy DM is an AI Game Master for a Fallout
+tabletop RPG (Modiphius 2d20). You talk to it; it narrates vivid scenes, voices the NPCs in
+distinct designed voices, rolls the dice, reads your physical character sheet through a camera,
+sets the mood with your room's smart lights, and remembers your party — all embodied by a
+**Reachy Mini** robot. Every model in the stack is open-weights and under the 32B cap.
+
+{SOCIAL_LINE}**Code:** [{GITHUB_URL.split('//')[-1]}]({GITHUB_URL}) (branch `cascade-integration`)
+
+> The full experience is a **local hardware rig** (robot · smart lights · companion screen),
+> so the **demo video above** is the showcase. This page explains how it works.
+
+## How it works — an all-Qwen cascade
+A voice loop (Whisper STT · Silero VAD) drives a **Qwen brain** that orchestrates tools and
+speaks back in designed character voices, reading the table with vision when you show it something.
+
+| Stage | Model | Where |
+|---|---|---|
+| Speech-to-text | Whisper | OpenAI API |
+| Voice activity | Silero VAD | local (CPU) |
+| **DM brain (logic + tool-calling)** | **Qwen3-30B-A3B-Instruct-2507 (FP8)** | **Modal** (1×H100) |
+| **Character voices (11 designed)** | **Qwen3-TTS-12Hz-1.7B** | **Modal** (L4) |
+| **Vision ("read the table")** | **Qwen3-VL-8B-Instruct** | **Modal** (L40S) |
+
+The brain calls tools over **MCP**: dice / character sheet / player choice render as live
+**Pip-Boy widgets** (an MCP-Apps custom UI), plus per-character voices (`speak_as`), durable
+**memory**, the camera, robot motion/expression, and Home Assistant smart lighting.
+
+### Best Use of Modal
+The three GPU models — **brain (H100)**, **character-voice TTS (L4)**, **vision (L40S)** — are all
+served on **Modal** (serverless vLLM + a custom FastAPI TTS server). Modal is the runtime
+compute backbone of the whole experience.
+
+### Badges
+- **Best Agent** — a genuine multi-step agent choosing among dice, sheets, choices, voices,
+  vision, memory, lighting, and robot motion, all under the 32B cap.
+- **Best Demo** — a robot Game Master that voices NPCs, rolls dice on screen, dims your lights
+  for combat, and reads your printed character sheet with a camera.
+- **Off Brand** — tool calls render as a custom **Pip-Boy** UI well past the default Gradio look.
+
+## Models & constraints
+All-Qwen, open-weights, **each model < 32B total** (30B brain · 8B vision · 1.7B TTS), plus
+Whisper/Silero for speech. Built on a fork of the Pollen Robotics Reachy Mini conversation app.
+
+## Disclaimer
+Reachy DM is a personal, non-commercial creative project built for the Build Small hackathon. It
+runs *Fallout: The Roleplaying Game* (Modiphius 2d20 system) using rules and content from a copy
+I own, as a fan would at their own table. It is **not affiliated with, endorsed by, or sponsored
+by Bethesda Softworks or Modiphius Entertainment**. *Fallout* is a trademark of Bethesda Softworks;
+*Fallout: The Roleplaying Game* is published by Modiphius Entertainment. All rights to those
+properties belong to their respective owners.
+"""
+
+with gr.Blocks(css=PIPBOY_CSS, title="Reachy DM — Thousand Token Wood") as demo:
+    with gr.Column(elem_id="pip"):
+        gr.HTML(VIDEO_BLOCK)
+        gr.Markdown(WRITEUP)
+
+if __name__ == "__main__":
+    demo.launch()

requirements.txt

@@ -0,0 +1 @@
+gradio>=5,<6