Zen4 Coder - rebranded from Qwen3-Coder-Next abliterated

README.md

@@ -1,241 +1,65 @@
 ---
-library_name: transformers
 license: apache-2.0
-license_link: https://huggingface.co/Qwen/Qwen3-Coder-Next/blob/main/LICENSE
-pipeline_tag: text-generation
-base_model:
-- Qwen/Qwen3-Coder-Next
+language:
+- en
+- zh
 tags:
+- zen4
+- zenlm
+- hanzo
 - abliterated
 - uncensored
+base_model: huihui-ai/Huihui-Qwen3-Coder-Next-abliterated
+pipeline_tag: text-generation
 ---
 
-# huihui-ai/Huihui-Qwen3-Coder-Next-abliterated
-
+# Zen4 Coder
 
-This is an uncensored version of [Qwen/Qwen3-Coder-Next](https://huggingface.co/Qwen/Qwen3-Coder-Next) created with abliteration (see [remove-refusals-with-transformers](https://github.com/Sumandora/remove-refusals-with-transformers) to know more about it).
-This is a crude, proof-of-concept implementation to remove refusals from an LLM model without using TransformerLens.
+**Zen4 Coder** is a 80B MoE (3B active) parameter language model from the [Zen4 family](https://zenlm.org) by [Zen LM](https://huggingface.co/zenlm) and [Hanzo AI](https://hanzo.ai).
 
-## ollama
-
-Please use the latest version of [ollama 0.15.5](https://github.com/ollama/ollama/releases/tag/v0.15.5)
-
-You can use [huihui_ai/qwen3-coder-next-abliterated](https://ollama.com/huihui_ai/qwen3-coder-next-abliterated) directly, 
-```
-ollama run huihui_ai/qwen3-coder-next-abliterated
-```
+Built on abliterated (uncensored) weights from Qwen3-Coder-Next for unrestricted, open-ended AI assistance.
 
-## chat_template-vl.jinja
+## Model Details
 
-We have added a new file named [chat_template-vl.jinja](https://huggingface.co/huihui-ai/Huihui-Qwen3-Coder-Next-abliterated/blob/main/chat_template-vl.jinja), which comes from the path `huihui-ai/Huihui-Qwen3-VL-30B-A3B-Instruct-abliterated`. 
+| Property | Value |
+|----------|-------|
+| **Parameters** | 80B MoE total, 3B active |
+| **Context** | 256K tokens |
+| **Base** | Qwen3-Coder-Next (abliterated) |
+| **License** | Apache-2.0 |
+| **Family** | Zen4 |
+| **Creator** | Zen LM / Hanzo AI |
 
-The new file chat_template-vl.jinja is more compatible with using Tool Calling in [llama-server](https://github.com/ggml-org/llama.cpp/releases/tag/b7952), 
-especially when [opencode](https://github.com/anomalyco/opencode/releases/tag/v1.1.53) is involved.
+## Zen4 Family
 
+| Model | Params | Active | Context | HuggingFace |
+|-------|--------|--------|---------|-------------|
+| Zen4 Mini | 4B | 4B | 32K | [zenlm/zen4-mini](https://huggingface.co/zenlm/zen4-mini) |
+| Zen4 | 8B | 8B | 32K | [zenlm/zen4](https://huggingface.co/zenlm/zen4) |
+| Zen4 Pro | 14B | 14B | 32K | [zenlm/zen4-pro](https://huggingface.co/zenlm/zen4-pro) |
+| Zen4 Max | 30B MoE | 3B | 256K | [zenlm/zen4-max](https://huggingface.co/zenlm/zen4-max) |
+| **Zen4 Pro Max** | **80B MoE** | **3B** | **256K** | [zenlm/zen4-pro-max](https://huggingface.co/zenlm/zen4-pro-max) |
+| Zen4 Coder Flash | 31B MoE | 3B | 131K | [zenlm/zen4-coder-flash](https://huggingface.co/zenlm/zen4-coder-flash) |
+| **Zen4 Coder** | **80B MoE** | **3B** | **256K** | [zenlm/zen4-coder](https://huggingface.co/zenlm/zen4-coder) |
+| Zen4 Ultra | 1.04T MoE | 32B | 256K | [zenlm/zen4-ultra](https://huggingface.co/zenlm/zen4-ultra) |
 
 ## Usage
-You can use this model in your applications by loading it with Hugging Face's `transformers` library:
 
 ```python
-from transformers import AutoModelForCausalLM, AutoTokenizer, TextStreamer, BitsAndBytesConfig
-import torch
-import os
-import signal
-import random
-import numpy as np
-import time
-import sys
-
-if (
-    "PYTORCH_ALLOC_CONF" not in os.environ
-    and "PYTORCH_CUDA_ALLOC_CONF" not in os.environ
-):
-    print(f"PYTORCH_ALLOC_CONF.")
-    os.environ["PYTORCH_ALLOC_CONF"] = "expandable_segments:True"
-
-cpu_count = os.cpu_count()
-print(f"Number of CPU cores in the system: {cpu_count}")
-half_cpu_count = cpu_count // 2
-os.environ["MKL_NUM_THREADS"] = str(half_cpu_count)
-os.environ["OMP_NUM_THREADS"] = str(half_cpu_count)
-torch.set_num_threads(half_cpu_count)
-
-print(f"PyTorch threads: {torch.get_num_threads()}")
-print(f"MKL threads: {os.getenv('MKL_NUM_THREADS')}")
-print(f"OMP threads: {os.getenv('OMP_NUM_THREADS')}")
-
-# Load the model and tokenizer
-MODEL_ID = "huihui-ai/Huihui-Qwen3-Coder-Next-abliterated"
-
-print(f"Load Model {MODEL_ID} ... ")
-quant_config_4 = BitsAndBytesConfig(
-    load_in_4bit=True,
-    bnb_4bit_compute_dtype=torch.bfloat16,
-    bnb_4bit_use_double_quant=True,
-    llm_int8_enable_fp32_cpu_offload=True,
-)
-
-model = AutoModelForCausalLM.from_pretrained(
-    MODEL_ID,
-    device_map="auto",
-    trust_remote_code=True,
-    torch_dtype="auto",
-    low_cpu_mem_usage=True,
-    quantization_config=quant_config_4,
-)
-
-tokenizer = AutoTokenizer.from_pretrained(MODEL_ID, trust_remote_code=True)
-
-messages = []
-skip_prompt=True
-skip_special_tokens=True
-
-class CustomTextStreamer(TextStreamer):
-    def __init__(self, tokenizer, skip_prompt=True, skip_special_tokens=True):
-        super().__init__(tokenizer, skip_prompt=skip_prompt, skip_special_tokens=skip_special_tokens)
-        self.generated_text = ""
-        self.stop_flag = False
-        self.init_time = time.time()  # Record initialization time
-        self.end_time = None  # To store end time
-        self.first_token_time = None  # To store first token generation time
-        self.token_count = 0  # To track total tokens
-
-    def on_finalized_text(self, text: str, stream_end: bool = False):
-        if self.first_token_time is None and text.strip():  # Set first token time on first non-empty text
-            self.first_token_time = time.time()
-        self.generated_text += text
-
-        self.token_count += 1
-
-        print(text, end="", flush=True)
-        if stream_end:
-            self.end_time = time.time()  # Record end time when streaming ends
-        if self.stop_flag:
-            raise StopIteration
-
-    def stop_generation(self):
-        self.stop_flag = True
-        self.end_time = time.time()  # Record end time when generation is stopped
-
-    def get_metrics(self):
-        """Returns initialization time, first token time, first token latency, end time, total time, total tokens, and tokens per second."""
-        if self.end_time is None:
-            self.end_time = time.time()  # Set end time if not already set
-        total_time = self.end_time - self.init_time  # Total time from init to end
-        tokens_per_second = self.token_count / total_time if total_time > 0 else 0
-        first_token_latency = (self.first_token_time - self.init_time) if self.first_token_time is not None else None
-        metrics = {
-            "init_time": self.init_time,
-            "first_token_time": self.first_token_time,
-            "first_token_latency": first_token_latency,
-            "end_time": self.end_time,
-            "total_time": total_time,  # Total time in seconds
-            "total_tokens": self.token_count,
-            "tokens_per_second": tokens_per_second
-        }
-        return metrics
-
-def generate_stream(model, tokenizer, messages, skip_prompt, skip_special_tokens, max_new_tokens):
-    text = tokenizer.apply_chat_template(
-        messages,
-        tokenize=False,
-        add_generation_prompt=True,
-    )
-    model_inputs = tokenizer(
-        [text],
-        return_tensors="pt",
-    ).to(model.device)
-
-    streamer = CustomTextStreamer(tokenizer, skip_prompt=skip_prompt, skip_special_tokens=skip_special_tokens)
+from transformers import AutoModelForCausalLM, AutoTokenizer
 
-    def signal_handler(sig, frame):
-        streamer.stop_generation()
-        print("\n[Generation stopped by user with Ctrl+C]")
+model = AutoModelForCausalLM.from_pretrained("zenlm/zen4-coder")
+tokenizer = AutoTokenizer.from_pretrained("zenlm/zen4-coder")
 
-    signal.signal(signal.SIGINT, signal_handler)
-
-    print("Response: ", end="", flush=True)
-    try:
-        generated_ids = model.generate(
-            **model_inputs,
-            max_new_tokens = max_new_tokens,
-            streamer=streamer,
-        )
-        del generated_ids
-    except StopIteration:
-        print("\n[Stopped by user]")
-
-    del model_inputs
-    torch.cuda.empty_cache()
-    signal.signal(signal.SIGINT, signal.SIG_DFL)
-
-    return streamer.generated_text, streamer.stop_flag, streamer.get_metrics()
-
-
-while True:
-    print(f"skip_prompt: {skip_prompt}")
-    print(f"skip_special_tokens: {skip_special_tokens}")
-
-    user_input = input("User: ").strip()
-    if user_input.lower() == "/exit":
-        print("Exiting chat.")
-        break
-    if user_input.lower() == "/clear":
-        messages = []
-        print("Chat history cleared. Starting a new conversation.")
-        continue
-    if user_input.lower() == "/skip_prompt":
-        skip_prompt = not skip_prompt
-        continue
-    if user_input.lower() == "/skip_special_tokens":
-        skip_special_tokens = not skip_special_tokens
-        continue
-    if not user_input:
-        print("Input cannot be empty. Please enter something.")
-        continue
-
-
-    messages.append({
-        "role": "user",
-        "content": user_input
-    })
-
-    response, stop_flag, metrics = generate_stream(model, tokenizer, messages, skip_prompt, skip_special_tokens, 40960)
-    print("\n\nMetrics:")
-    for key, value in metrics.items():
-        print(f"  {key}: {value}")
-
-
-    print("", flush=True)
-    if stop_flag:
-        continue
-    messages.append({
-        "role": "assistant",
-        "content": response.strip()
-    })
+messages = [{"role": "user", "content": "Hello, who are you?"}]
+text = tokenizer.apply_chat_template(messages, tokenize=False, add_generation_prompt=True)
+inputs = tokenizer(text, return_tensors="pt")
+outputs = model.generate(**inputs, max_new_tokens=512)
+print(tokenizer.decode(outputs[0], skip_special_tokens=True))
 ```
 
-### Usage Warnings
-
-
- - **Risk of Sensitive or Controversial Outputs**: This model’s safety filtering has been significantly reduced, potentially generating sensitive, controversial, or inappropriate content. Users should exercise caution and rigorously review generated outputs.
-
- - **Not Suitable for All Audiences**: Due to limited content filtering, the model’s outputs may be inappropriate for public settings, underage users, or applications requiring high security.
-
- - **Legal and Ethical Responsibilities**: Users must ensure their usage complies with local laws and ethical standards. Generated content may carry legal or ethical risks, and users are solely responsible for any consequences.
-
- - **Research and Experimental Use**: It is recommended to use this model for research, testing, or controlled environments, avoiding direct use in production or public-facing commercial applications.
-
- - **Monitoring and Review Recommendations**: Users are strongly advised to monitor model outputs in real-time and conduct manual reviews when necessary to prevent the dissemination of inappropriate content.
-
- - **No Default Safety Guarantees**: Unlike standard models, this model has not undergone rigorous safety optimization. huihui.ai bears no responsibility for any consequences arising from its use.
-
-
-### Donation
-##### Your donation helps us continue our further development and improvement, a cup of coffee can do it.
-- bitcoin:
-```
-  bc1qqnkhuchxw0zqjh2ku3lu4hq45hc6gy84uk70ge
-```
-- Support our work on [Ko-fi](https://ko-fi.com/huihuiai)!
+## Links
 
+- [Zen LM](https://zenlm.org) | [Hanzo AI](https://hanzo.ai)
+- [GitHub](https://github.com/zenlm/zen4-coder)
+- [All Zen4 Models](https://huggingface.co/zenlm)