Expand model card

README.md

@@ -21,7 +21,7 @@ inference: false
 
 ### The flagship. Strict zero retention.
 
-![interface](https://img.shields.io/badge/interface-OpenAI--compatible-black) ![license](https://img.shields.io/badge/license-Apache--2.0-green) ![skeleton-v1](https://img.shields.io/badge/zero%20data%20retention-purple) ![benchmarks](https://img.shields.io/badge/benchmarks-self--reported-orange)
+![interface](https://img.shields.io/badge/interface-OpenAI--compatible-black) ![license](https://img.shields.io/badge/license-Apache--2.0-green) ![skeleton-v1](https://img.shields.io/badge/zero%20data%20retention-purple) ![context](https://img.shields.io/badge/context-128K-blue) ![benchmarks](https://img.shields.io/badge/benchmarks-self--reported-orange)
 
 **Stealth-Skeleton Research (`stealth-skeleton`)**
 
@@ -36,26 +36,64 @@ inference: false
 
 > **API-only.** This repository is the official Model Card & Documentation Hub for the `stealth-skeleton/skeleton-v1` endpoint on OpenRouter. It does **not** contain downloadable model weights. The Apache-2.0 license applies to the documentation and configuration files in this repository.
 
-**Flagship Sovereign Mathematical, Spectral & Systems Reasoning Model**
+## 👑 Overview: Sovereign General-Purpose Mathematical, Systems & Code Reasoning Flagship
 
-The flagship of the family: long-form mathematical, spectral and systems reasoning with a strict zero-data-retention policy.
+Full 3-pass inference stack (Pass 1 fast draft → Pass 2 chain-of-thought critique → Pass 3 operator-verified synthesis) with strict **Zero Data Retention (ZDR)**. Also supports `service_tier: "priority"` (`skeleton-v1-fast`) and `variant: "batch"`.
 
-## ✨ What it does
+- **Full 3-pass stack**: draft, critique, operator-verified synthesis.
+- **32K max output** for full derivations and multi-file work.
+- **Strict ZDR**, priority tier and batch variant available.
 
-- **Sovereign reasoning** over mathematical, spectral and systems problems.
-- **32K max output** for full derivations and long-form answers.
-- **Strict Zero Data Retention (ZDR)**: ephemeral processing only.
+**Best for:**
+
+- Production software engineering & complex multi-file refactoring
+- Enterprise workloads requiring strict **Zero Data Retention (ZDR)**
+- Long-horizon system design, formal verification & deep research synthesis
 
 ## 🧾 Quick facts
 
 | Property | Value |
 |---|---|
-| Interface | OpenAI-compatible Chat Completions |
+| Tier | Flagship Production |
+| Price (input / output per 1M tokens) | $1.50 / $6.00 |
+| Cache read | $0.30 per 1M |
 | Context window | 128K tokens |
 | Max output | 32K tokens |
-| Pricing | See the OpenRouter listing |
+| Data policy | Strict ZDR |
+| Interface | OpenAI-compatible Chat Completions |
 | Weights | None: API-only |
 
+## 🦴 Stealth-Skeleton Model Family Guide: What Each Model Does & When to Use It
+
+| Model | Tier, pricing & limits | Core architecture & what it does | Best use cases (when to use it) |
+|---|---|---|---|
+| 💀 **[skeleton-alpha](https://huggingface.co/stealth-skeleton/skeleton-alpha)** | **Free Preview**<br>$0 / $0 (free) per 1M<br>128K context, 16K max output | Combines multi-pass reasoning with **Auto-Exacto** deterministic tool-call grammar healing (>95% schema compliance, self-reported) and symbolic conservation checks. Anonymized execution-verified traces help align future checkpoints via DPO/GRPO. | • Daily agentic coding in **OpenCode, Cline, Roo Code, Cursor, Aider**<br>• Multi-turn function/tool calling & MCP orchestration<br>• Zero-cost high-volume prototyping, unit test generation & math/code verification |
+| 👑 **[skeleton-v1](https://huggingface.co/stealth-skeleton/skeleton-v1)** | **Flagship Production**<br>$1.50 / $6.00 per 1M<br>128K context, 32K max output | Full 3-pass inference stack (Pass 1 fast draft → Pass 2 chain-of-thought critique → Pass 3 operator-verified synthesis) with strict **Zero Data Retention (ZDR)**. Also supports `service_tier: "priority"` (`skeleton-v1-fast`) and `variant: "batch"`. | • Production software engineering & complex multi-file refactoring<br>• Enterprise workloads requiring strict **Zero Data Retention (ZDR)**<br>• Long-horizon system design, formal verification & deep research synthesis |
+| 📐 **[skeleton-math-pro](https://huggingface.co/stealth-skeleton/skeleton-math-pro)** | **Specialist Kernel**<br>$1.50 / $6.00 per 1M<br>128K context, 32K max output | Built for symbolic derivations, **Sinkhorn-Knopp optimal transport** (L1 marginal residual 1.39e-16), **Frisch-Waugh-Lovell (FWL)** orthogonal residualization (< 1e-15), **36th-order Hou-Li spectral filtering**, and Lyapunov stability proofs. | • Graduate-level mathematics, physics, quantitative finance & operations research<br>• Convex optimization, optimal transport, spectral graph theory & PDE/SDE modeling<br>• Verifying symbolic equations and numerical stability invariants |
+| 🔮 **[skeleton-forecast](https://huggingface.co/stealth-skeleton/skeleton-forecast)** | **Specialist Forecaster**<br>$1.50 / $6.00 per 1M<br>128K context, 32K max output | Applies **Romano Conformalized Quantile Regression (CQR)** (90.35% empirical coverage), **Gibbs-Candes Adaptive Conformal Inference (ACI)**, **Barber non-exchangeable conformal weights**, **Kadavath P(IK) epistemic shrinkage**, and **Satopaa-Tetlock log-odds extremizing** (0.0969 ForecastBench Brier). | • Calibrated probability estimation & risk modeling where raw LLMs are overconfident<br>• Time-series quantile intervals (80% / 90% conformal bounds) under distribution shift<br>• Quantitative decision support, calibration & proper scoring rule (Brier / CRPS) tasks |
+| 🗣️ **[skeleton-council](https://huggingface.co/stealth-skeleton/skeleton-council)** | **Multi-Agent Arbiter**<br>$1.50 / $6.00 per 1M<br>128K context, 32K max output | Runs four orthogonal internal perspectives (**Architect**, **Skeptic**, **Pragmatist**, **Critic**) reconciled via **Cellular Sheaf Laplacian cohomology** and an **80-rule misleading-heuristic firewall** to reduce sycophancy and groupthink. | • High-stakes architecture reviews, security audits & go/no-go engineering decisions<br>• Red-teaming plans, detecting spurious correlations & stress-testing assumptions<br>• Final conjunction-gate arbitration before deploying code or capital |
+| ⚡ **[skeleton-flash](https://huggingface.co/stealth-skeleton/skeleton-flash)** | **Low-Latency / High-Throughput**<br>$0.25 / $1.00 per 1M<br>128K context, 16K max output | Optimized for ultra-fast Time-to-First-Token (TTFT), hierarchical document tree traversal without lossy vector-chunk truncation, and high-concurrency extraction. | • Real-time chat, sub-second routing, classification & document extraction<br>• Large-corpus hierarchical RAG over books, PDFs, and codebases<br>• High-concurrency batch pipelines & subagent worker loops |
+
+### 🧭 Quick decision guide: which model should I pick?
+
+1. **Want a free daily-driver for coding, agents, or tool use?** → use 💀 [`stealth-skeleton/skeleton-alpha`](https://huggingface.co/stealth-skeleton/skeleton-alpha)
+2. **Need flagship coding + reasoning with strict Zero Data Retention (ZDR)?** → use 👑 [`stealth-skeleton/skeleton-v1`](https://huggingface.co/stealth-skeleton/skeleton-v1)
+3. **Solving hard proofs, optimal transport, linear algebra, PDEs, or numerical optimization?** → use 📐 [`stealth-skeleton/skeleton-math-pro`](https://huggingface.co/stealth-skeleton/skeleton-math-pro)
+4. **Need calibrated probabilities, Brier/CRPS optimization, or conformal prediction intervals?** → use 🔮 [`stealth-skeleton/skeleton-forecast`](https://huggingface.co/stealth-skeleton/skeleton-forecast)
+5. **Need an adversarial 4-voice design/security review that will not rubber-stamp your idea?** → use 🗣️ [`stealth-skeleton/skeleton-council`](https://huggingface.co/stealth-skeleton/skeleton-council)
+6. **Need maximum speed and low cost for RAG, extraction, or high-volume worker loops?** → use ⚡ [`stealth-skeleton/skeleton-flash`](https://huggingface.co/stealth-skeleton/skeleton-flash)
+
+### 👀 The whole family at a glance
+
+| Model | In one line |
+|---|---|
+| 💀 [skeleton-alpha](https://huggingface.co/stealth-skeleton/skeleton-alpha) | Free. Fast. Execution-verified. |
+| 👑 **skeleton-v1** 👈 you are here | The flagship. Strict zero retention. |
+| 📐 [skeleton-math-pro](https://huggingface.co/stealth-skeleton/skeleton-math-pro) | Graduate math, spectral filters, transport. |
+| 🔮 [skeleton-forecast](https://huggingface.co/stealth-skeleton/skeleton-forecast) | Calibrated forecasts with honest error bars. |
+| 🗣️ [skeleton-council](https://huggingface.co/stealth-skeleton/skeleton-council) | Four voices argue. A sheaf decides. |
+| ⚡ [skeleton-flash](https://huggingface.co/stealth-skeleton/skeleton-flash) | Quick retrieval. No vector database. |
+
 ## 🔬 Under the hood
 
 - **Execution-verified mathematical operator synthesis**: generated operators are checked by running them.
@@ -65,12 +103,19 @@ The flagship of the family: long-form mathematical, spectral and systems reasoni
 - **36th-order Hou-Li spectral filtering**.
 - **Auto-Exacto tool-calling grammar healing**: malformed tool-call JSON is repaired before it is returned.
 
+### The 3-pass stack
+
+```text
+ Pass 1            Pass 2                 Pass 3
+ fast draft  --->  chain-of-thought  ---> operator-verified
+                   critique               synthesis
+ 43.75             78.52                  86.12   (composite, self-reported)
+```
+
 All figures above are self-reported from internal tests.
 
 ## 📊 Benchmarks: Internal 3-Pass Evaluation Suite (Self-Reported)
 
-These results come from our internal 3-pass evaluation suite. They have not been independently verified and may differ from official leaderboard runs.
-
 | Suite | Score |
 |---|---|
 | MATH-500 | 92.00% |
@@ -82,12 +127,25 @@ These results come from our internal 3-pass evaluation suite. They have not been
 | ForecastBench (Proper Brier Skill) | 78.52% (0.0969 Brier) |
 | **Composite (mean of 7)** | **86.12 / 100** |
 
-## 🚀 Try it in 30 seconds
+### 🥊 Self-Reported 3-Pass Evaluation vs Published Peer Baselines
+
+| Model | Composite (100-pt) | MATH-500 | GSM8K | HumanEval | GPQA-Diamond | MMLU-Pro | ForecastBench Brier (lower is better) | API Pricing (In / Out per 1M) |
+|---|---:|---:|---:|---:|---:|---:|---:|---:|
+| **stealth-skeleton/skeleton-v1 (3-Pass, Self-Reported)** | **86.12** | **92.0%** (85.0% slice) | **96.0%** (95.0% slice) | **92.0%** | **73.3%** (75.0% slice) | **86.0%** | **0.0969** | **$1.50 / $6.00** (skeleton-alpha: **Free**) |
+| OpenAI o1 (Published Reference) | 87.54 | 94.8% | 96.4% | 92.4% | 75.7% | 81.0% | 0.1010 | $15.00 / $60.00 |
+| DeepSeek-R1 (Published Reference) | 86.40 | 97.3% | 96.1% | 92.1% | 71.5% | 84.0% | 0.1090 | $0.55 / $2.19 |
+| Claude 3.5 Sonnet (Oct, Published Reference) | 80.55 | 78.0% | 96.4% | 93.7% | 65.0% | 78.0% | 0.1040 | $3.00 / $15.00 |
+| DeepSeek-V3 (Pass 2 Reference) | 78.52 | 90.2% | 90.0% | 89.0% | 59.1% | 75.9% | 0.1140 | $0.27 / $1.10 |
+| GPT-4o (Published Reference) | 77.90 | 76.6% | 95.8% | 90.2% | 53.6% | 74.4% | 0.1180 | $2.50 / $10.00 |
+| Qwen-2.5-7B (Pass 1 0-Shot Baseline) | 43.75 | 40.0% | 20.0% | 100.0% (20-item) | 25.0% | 55.0% (MMLU) | 0.2251 | $0.05 / $0.10 |
 
-Model ID: `stealth-skeleton/skeleton-v1` via OpenRouter, once provider routing is active for your account.
+*How to read this:* the skeleton rows are **self-reported** from our internal 3-pass evaluation (N=145 external questions plus N=403 resolved ForecastBench questions, Wilson 95% confidence intervals). Peer rows are published reference values, or our own reference runs where labeled. Composite scores use our internal 100-point weighting and are not published by the peer vendors. Nothing here has been independently verified: please run your own evaluations. For Brier, lower is better. Prices are OpenRouter-listed prices at the time of writing and can change.
 
-<details open>
-<summary><b>Python (OpenAI SDK)</b></summary>
+## 🚀 How to use the stealth-skeleton models
+
+Model IDs: `stealth-skeleton/skeleton-alpha`, `skeleton-v1`, `skeleton-math-pro`, `skeleton-forecast`, `skeleton-council`, `skeleton-flash` via OpenRouter, once provider routing is active for your account. Swap the model ID in any example.
+
+### 1. Python (openai SDK via OpenRouter)
 
 ```python
 from openai import OpenAI
@@ -97,32 +155,87 @@ client = OpenAI(
     api_key="<YOUR_OPENROUTER_API_KEY>",
 )
 
-resp = client.chat.completions.create(
+response = client.chat.completions.create(
     model="stealth-skeleton/skeleton-v1",
-    messages=[{"role": "user", "content": "Derive the stability condition for an explicit finite-difference scheme for the 1D heat equation."}],
+    messages=[
+        {"role": "system", "content": "You are a rigorous mathematical and systems engineering assistant."},
+        {"role": "user", "content": "Review this system design for failure modes and propose a verified refactoring plan."},
+    ],
+    temperature=0.2,
+    max_tokens=4096,
+    stream=False,
 )
-print(resp.choices[0].message.content)
+
+print(response.choices[0].message.content)
 ```
 
-</details>
+### 2. Tool calling (Auto-Exacto grammar healing)
 
-<details>
-<summary><b>curl</b></summary>
+```python
+tools = [
+    {
+        "type": "function",
+        "function": {
+            "name": "verify_conservation_law",
+            "description": "Verify L1 marginal residual on a transport matrix.",
+            "parameters": {
+                "type": "object",
+                "properties": {
+                    "matrix_dim": {"type": "integer"},
+                    "epsilon": {"type": "number"},
+                },
+                "required": ["matrix_dim", "epsilon"],
+            },
+        },
+    }
+]
+
+tool_resp = client.chat.completions.create(
+    model="stealth-skeleton/skeleton-v1",
+    messages=[{"role": "user", "content": "Run a 40-step conservation check on a 16x16 matrix with epsilon=0.15."}],
+    tools=tools,
+    tool_choice="auto",
+)
+print(tool_resp.choices[0].message.tool_calls)
+```
+
+### 3. TypeScript / Node.js
+
+```typescript
+import OpenAI from "openai";
+
+const client = new OpenAI({
+  baseURL: "https://openrouter.ai/api/v1",
+  apiKey: process.env.OPENROUTER_API_KEY,
+});
+
+const completion = await client.chat.completions.create({
+  model: "stealth-skeleton/skeleton-v1",
+  messages: [{ role: "user", content: "Review this system design for failure modes and propose a verified refactoring plan." }],
+});
+
+console.log(completion.choices[0]?.message?.content);
+```
+
+### 4. cURL
 
 ```bash
 curl https://openrouter.ai/api/v1/chat/completions \
-  -H "Authorization: Bearer $OPENROUTER_API_KEY" \
   -H "Content-Type: application/json" \
+  -H "Authorization: Bearer <YOUR_OPENROUTER_API_KEY>" \
   -d '{
     "model": "stealth-skeleton/skeleton-v1",
-    "messages": [{"role": "user", "content": "Hello, skeleton!"}]
+    "messages": [
+      {"role": "user", "content": "Hello, skeleton!"}
+    ],
+    "temperature": 0.1
   }'
 ```
 
-</details>
+### 5. OpenCode, Cline, Roo Code, Kilo Code & Aider
 
 <details>
-<summary><b>sst/opencode</b></summary>
+<summary><b>OpenCode (opencode.json)</b></summary>
 
 ```json
 {
@@ -139,16 +252,21 @@ curl https://openrouter.ai/api/v1/chat/completions \
 
 </details>
 
-## 🦴 Meet the family
+<details>
+<summary><b>Cline / Roo Code / Kilo Code</b></summary>
 
-| Model | In one line |
-|---|---|
-| 💀 [skeleton-alpha](https://huggingface.co/stealth-skeleton/skeleton-alpha) | Free. Fast. Execution-verified. |
-| 👑 [skeleton-v1](https://huggingface.co/stealth-skeleton/skeleton-v1) | The flagship. Strict zero retention. |
-| 📐 [skeleton-math-pro](https://huggingface.co/stealth-skeleton/skeleton-math-pro) | Graduate-level math, spectral filters, transport. |
-| 🔮 [skeleton-forecast](https://huggingface.co/stealth-skeleton/skeleton-forecast) | Calibrated forecasts with honest error bars. |
-| 🗣️ [skeleton-council](https://huggingface.co/stealth-skeleton/skeleton-council) | Four voices argue. A sheaf decides. |
-| ⚡ [skeleton-flash](https://huggingface.co/stealth-skeleton/skeleton-flash) | Quick retrieval. No vector database. |
+Set the API Provider to **OpenRouter**, enter your OpenRouter key, and pick `stealth-skeleton/skeleton-v1` as the model.
+
+</details>
+
+<details>
+<summary><b>Aider</b></summary>
+
+```bash
+aider --model openrouter/stealth-skeleton/skeleton-v1
+```
+
+</details>
 
 ## 🔒 Data & Privacy
 
@@ -156,6 +274,6 @@ curl https://openrouter.ai/api/v1/chat/completions \
 
 ## ⚠️ Limitations
 
-- Behavior, limits and availability may change without notice.
+- Behavior, limits, prices and availability may change without notice.
 - Benchmarks are self-reported only; validate on your own workloads.
 - Not intended for safety-critical, medical, legal or financial decisions.

config.json

@@ -4,7 +4,13 @@
   "organization": "Stealth-Skeleton Research",
   "interface": "openai-compatible-chat-completions",
   "weights_available": false,
+  "tier": "Flagship Production",
   "context_length": 131072,
   "max_output_tokens": 32768,
+  "pricing_usd_per_1m_tokens": {
+    "input": 1.5,
+    "output": 6,
+    "cache_read": 0.3
+  },
   "data_policy": "zero-data-retention"
 }