Enrich org card: pricing table, usage examples, model browser link (patterns from Groq/DeepInfra/Cerebras provider pages)

README.md

@@ -9,29 +9,69 @@ pinned: false
 
 # IntelCS-AI
 
-We run an **inference provider** on the Hugging Face Inference Providers
-platform: open-weight models served on rented GPU capacity, sold at
-marketplace prices.
+**Low-cost inference for open-weight models.** We serve open models on
+efficient GPU capacity at some of the lowest per-token prices on the
+platform — with no compromise on features.
 
-## What we offer
+> Browse [models served by IntelCS-AI on Hugging Face](https://huggingface.co/models?inference_provider=intelcs-ai&sort=trending)
 
-- **OpenAI-compatible endpoints** (`/v1/chat/completions`, `/v1/models`)
-  behind the HF Inference Providers API.
-- **Tool calling** (function calling) and **structured output**
-  (`response_format: json_schema`) support.
-- **Low-latency serving**: time-to-first-token well under the 5 s budget
+## Why IntelCS-AI
+
+- **Price-first**: open-weight models at floor prices — see the table below.
+- **Full feature parity**: tool calling (function calling) and structured
+  output (`response_format: json_schema`) on every conversational model.
+- **Long context**: up to 131K tokens of context on supported models.
+- **Low latency**: time-to-first-token well under the 5 s provider budget
   (measured ~0.9 s non-streaming).
 - **Autoscaling fleet**: capacity scales out automatically with demand;
   routing, metering, and billing run on edge infrastructure.
 
-## Currently serving
+## Pricing
+
+| Model | Input (per 1M tokens) | Output (per 1M tokens) | Context |
+|---|---|---|---|
+| `google/gemma-3-4b-it` | $0.05 | $0.10 | 131K |
+
+Our lineup rotates as we add capacity — the freshest list is always on the
+[models page](https://huggingface.co/models?inference_provider=intelcs-ai&sort=trending).
+
+## Usage
+
+OpenAI-compatible, through the standard Hugging Face clients:
+
+```python
+from huggingface_hub import InferenceClient
+
+client = InferenceClient(
+    model="google/gemma-3-4b-it",
+    provider="intelcs-ai",
+)
+
+# Chat with tool calling
+response = client.chat.completions.create(
+    messages=[{"role": "user", "content": "What's the weather in Berlin?"}],
+    tools=[{
+        "type": "function",
+        "function": {
+            "name": "get_weather",
+            "parameters": {"type": "object", "properties": {"location": {"type": "string"}}},
+        },
+    }],
+)
+print(response.choices[0].message.tool_calls)
 
-| Model | Notes |
-|---|---|
-| `google/gemma-3-4b-it` | 131K context, tool calling, JSON schema output |
+# Structured output
+response = client.chat.completions.create(
+    messages=[{"role": "user", "content": "Extract the city: 'Flight to Berlin delayed'"}],
+    response_format={"type": "json_schema", "json_schema": {"name": "city", "schema": {
+        "type": "object", "properties": {"city": {"type": "string"}}, "required": ["city"],
+    }}},
+)
+print(response.choices[0].message.content)  # {"city": "Berlin"}
+```
 
-Model lineup rotates as we add capacity — watch this space.
+Streaming (`stream=True`) is fully supported and metered per token.
 
-## Links
+## Resources
 
-- Website: [intelcs.ai](https://intelcs.ai)
+- **Website**: [intelcs.ai](https://intelcs.ai)