Update model card with Zen branding

README.md

@@ -1,55 +1,55 @@
 ---
-license: mit
-language:
-- en
-- zh
-tags:
-- zen4
-- zenlm
-- hanzo
-- abliterated
-- uncensored
-base_model: huihui-ai/Huihui-GLM-4.7-Flash-abliterated
+library_name: transformers
 pipeline_tag: text-generation
+tags:
+  - text-generation
+  - code
+  - zen
+  - zen4
+  - hanzo
+license: apache-2.0
 ---
 
 # Zen4 Coder Flash
 
-**31B MoE total, 3B active | 131K Context**
-
-Code-focused MoE model based on GLM-4.7-Flash with abliterated weights. 131K context for entire codebases. Tool calling and reasoning mode support.
-
-Part of the [Zen4 family](https://zenlm.org) by [Zen LM](https://huggingface.co/zenlm) and [Hanzo AI](https://hanzo.ai).
+**Zen LM by Hanzo AI** — Ultra-fast code generation model. 30B parameter MoE optimized for real-time coding assistance.
 
-## Model Details
+## Specs
 
 | Property | Value |
 |----------|-------|
-| **Parameters** | 31B MoE total, 3B active |
-| **Context** | 131K tokens |
-| **Base** | GLM-4.7-Flash (abliterated) |
-| **License** | MIT |
-| **Family** | Zen4 |
+| Parameters | 30B MoE |
+| Context Length | 131K tokens |
+| Architecture | Zen MoDE (Mixture of Distilled Experts) |
+| Generation | Zen4 |
+
+## API Access
+
+This model is served via the Hanzo AI API. Weight downloads are not available for this model tier.
+
+```bash
+curl https://api.hanzo.ai/v1/chat/completions \
+  -H "Authorization: Bearer $HANZO_API_KEY" \
+  -H "Content-Type: application/json" \
+  -d '{
+    "model": "zen4-coder-flash",
+    "messages": [{"role": "user", "content": "Hello"}],
+    "max_tokens": 1024
+  }'
+```
 
-## Usage
+Get your API key at [console.hanzo.ai](https://console.hanzo.ai) — $5 free credit on signup.
 
-```python
-from transformers import AutoModelForCausalLM, AutoTokenizer
+## Pricing
 
-model = AutoModelForCausalLM.from_pretrained("zenlm/zen4-coder-flash", torch_dtype="auto", device_map="auto")
-tokenizer = AutoTokenizer.from_pretrained("zenlm/zen4-coder-flash")
+| Tier | Input | Output |
+|------|-------|--------|
+| Standard | See [pricing](https://zenlm.org) | See [pricing](https://zenlm.org) |
 
-messages = [{"role": "user", "content": "Hello"}]
-text = tokenizer.apply_chat_template(messages, tokenize=False, add_generation_prompt=True)
-inputs = tokenizer([text], return_tensors="pt").to(model.device)
-outputs = model.generate(**inputs, max_new_tokens=512)
-print(tokenizer.decode(outputs[0], skip_special_tokens=True))
-```
+## License
 
-## About Zen4
+Apache 2.0
 
-The Zen4 family provides abliterated (uncensored) variants of the best open-source models, ranging from 4B to 1T+ parameters.
+---
 
-- **Website**: [zenlm.org](https://zenlm.org)
-- **Organization**: [Zen LM](https://huggingface.co/zenlm)
-- **Built by**: [Hanzo AI](https://hanzo.ai) (Techstars '17)
+*Zen LM is developed by [Hanzo AI](https://hanzo.ai) — Frontier AI infrastructure.*