Update README.md

README.md

@@ -1,351 +1,152 @@
----
-library_name: transformers
-license: apache-2.0
-license_link: https://huggingface.co/Qwen/Qwen3-32B/blob/main/LICENSE
-pipeline_tag: text-generation
----
+# Introduction
 
-# Qwen3-32B
-<a href="https://chat.qwen.ai/" target="_blank" style="margin: 2px;">
-    <img alt="Chat" src="https://img.shields.io/badge/%F0%9F%92%9C%EF%B8%8F%20Qwen%20Chat%20-536af5" style="display: inline-block; vertical-align: middle;"/>
-</a>
+**FlagOS** is a unified heterogeneous computing software stack for large models, co-developed with leading global chip manufacturers. With core technologies such as the **FlagScale** distributed training/inference framework, **FlagGems** universal operator library, **FlagCX** communication library, and **FlagTree** unified compiler, the **FlagRelease** platform leverages the FlagOS stack to automatically produce and release various combinations of <chip + open-source model>. This enables efficient and automated model migration across diverse chips, opening a new chapter for large model deployment and application.
 
-## Qwen3 Highlights
+Based on this, the **Qwen3-32B-FlagOS** model is adapted for the Nvidia chip using the FlagOS software stack, enabling:
 
-Qwen3 is the latest generation of large language models in Qwen series, offering a comprehensive suite of dense and mixture-of-experts (MoE) models. Built upon extensive training, Qwen3 delivers groundbreaking advancements in reasoning, instruction-following, agent capabilities, and multilingual support, with the following key features:
+### Integrated Deployment
 
-- **Uniquely support of seamless switching between thinking mode** (for complex logical reasoning, math, and coding) and **non-thinking mode** (for efficient, general-purpose dialogue) **within single model**, ensuring optimal performance across various scenarios.
-- **Significantly enhancement in its reasoning capabilities**, surpassing previous QwQ (in thinking mode) and Qwen2.5 instruct models (in non-thinking mode) on mathematics, code generation, and commonsense logical reasoning.
-- **Superior human preference alignment**, excelling in creative writing, role-playing, multi-turn dialogues, and instruction following, to deliver a more natural, engaging, and immersive conversational experience.
-- **Expertise in agent capabilities**, enabling precise integration with external tools in both thinking and unthinking modes and achieving leading performance among open-source models in complex agent-based tasks.
-- **Support of 100+ languages and dialects** with strong capabilities for **multilingual instruction following** and **translation**.
+- Deep integration with the open-source [FlagScale framework](https://github.com/FlagOpen/FlagScale)
+- Out-of-the-box inference scripts with pre-configured hardware and software parameters
+- Released **FlagOS** container image supporting deployment within minutes
 
-## Model Overview
+### Consistency Validation
 
-**Qwen3-32B** has the following features:
-- Type: Causal Language Models
-- Training Stage: Pretraining & Post-training
-- Number of Parameters: 32.8B
-- Number of Paramaters (Non-Embedding): 31.2B
-- Number of Layers: 64
-- Number of Attention Heads (GQA): 64 for Q and 8 for KV
-- Context Length: 32,768 natively and [131,072 tokens with YaRN](#processing-long-texts). 
+- Rigorously evaluated through benchmark testing: Performance and results from the FlagOS software stack are compared against native stacks on multiple public.
 
-For more details, including benchmark evaluation, hardware requirements, and inference performance, please refer to our [blog](https://qwenlm.github.io/blog/qwen3/), [GitHub](https://github.com/QwenLM/Qwen3), and [Documentation](https://qwen.readthedocs.io/en/latest/).
+# Technical Overview
 
-## Quickstart
+## **FlagScale Distributed Training and Inference Framework**
 
-The code of Qwen3 has been in the latest Hugging Face `transformers` and we advise you to use the latest version of `transformers`.
+FlagScale is an end-to-end framework for large models across heterogeneous computing resources, maximizing computational efficiency and ensuring model validity through core technologies. Its key advantages include:
 
-With `transformers<4.51.0`, you will encounter the following error:
-```
-KeyError: 'qwen3'
-```
+- **Unified Deployment Interface:** Standardized command-line tools support one-click service deployment across multiple hardware platforms, significantly reducing adaptation costs in heterogeneous environments.
+- **Intelligent Parallel Optimization:** Automatically generates optimal distributed parallel strategies based on chip computing characteristics, achieving dynamic load balancing of computation/communication resources.
+- **Seamless Operator Switching:** Deep integration with the FlagGems operator library allows high-performance operators to be invoked via environment variables without modifying model code.
 
-The following contains a code snippet illustrating how to use the model generate content based on given inputs. 
-```python
-from transformers import AutoModelForCausalLM, AutoTokenizer
+## **FlagGems Universal Large-Model Operator Library**
 
-model_name = "Qwen/Qwen3-32B"
+FlagGems is a Triton-based, cross-architecture operator library collaboratively developed with industry partners. Its core strengths include:
 
-# load the tokenizer and the model
-tokenizer = AutoTokenizer.from_pretrained(model_name)
-model = AutoModelForCausalLM.from_pretrained(
-    model_name,
-    torch_dtype="auto",
-    device_map="auto"
-)
+- **Full-stack Coverage**: Over 100 operators, with a broader range of operator types than competing libraries.
+- **Ecosystem Compatibility**: Supports 7 accelerator backends. Ongoing optimizations have significantly improved performance.
+- **High Efficiency**: Employs unique code generation and runtime optimization techniques for faster secondary development and better runtime performance compared to alternatives.
 
-# prepare the model input
-prompt = "Give me a short introduction to large language model."
-messages = [
-    {"role": "user", "content": prompt}
-]
-text = tokenizer.apply_chat_template(
-    messages,
-    tokenize=False,
-    add_generation_prompt=True,
-    enable_thinking=True # Switches between thinking and non-thinking modes. Default is True.
-)
-model_inputs = tokenizer([text], return_tensors="pt").to(model.device)
+## **FlagEval Evaluation Framework**
 
-# conduct text completion
-generated_ids = model.generate(
-    **model_inputs,
-    max_new_tokens=32768
-)
-output_ids = generated_ids[0][len(model_inputs.input_ids[0]):].tolist() 
+FlagEval (Libra)** is a comprehensive evaluation system and open platform for large models launched in 2023. It aims to establish scientific, fair, and open benchmarks, methodologies, and tools to help researchers assess model and training algorithm performance. It features:
+  - **Multi-dimensional Evaluation**: Supports 800+ model evaluations across NLP, CV, Audio, and Multimodal fields, covering 20+ downstream tasks including language understanding and image-text generation.
+  - **Industry-Grade Use Cases**: Has completed horizontal evaluations of mainstream large models, providing authoritative benchmarks for chip-model performance validation.
 
-# parsing thinking content
-try:
-    # rindex finding 151668 (</think>)
-    index = len(output_ids) - output_ids[::-1].index(151668)
-except ValueError:
-    index = 0
+# Evaluation Results
 
-thinking_content = tokenizer.decode(output_ids[:index], skip_special_tokens=True).strip("\n")
-content = tokenizer.decode(output_ids[index:], skip_special_tokens=True).strip("\n")
+## Benchmark Result 
 
-print("thinking content:", thinking_content)
-print("content:", content)
-```
+| Metrics           | Qwen3-32B-H100-CUDA | Qwen3-32B-FlagOS |
+|-------------------|--------------------------|-----------------------------|
+| AIME_0fewshot_@avg1 | 0.800 | 0.800 |
+| GPQA_0fewshot_@avg1 | 0.608 | 0.612 |
+| LiveBench-0fewshot_@avg1 | 0.591 | 0.568 |
+| MMLU_5fewshot_@avg1 | 0.770 | 0.769 |
+| MUSR_0fewshot_@avg | 0.644 | 0.673 |
 
-For deployment, you can use `sglang>=0.4.6.post1` or `vllm>=0.8.5` or to create an OpenAI-compatible API endpoint:
-- SGLang:
-    ```shell
-    python -m sglang.launch_server --model-path Qwen/Qwen3-32B --reasoning-parser qwen3
-    ```
-- vLLM:
-    ```shell
-    vllm serve Qwen/Qwen3-32B --enable-reasoning --reasoning-parser deepseek_r1
-    ```
+# User Guide
 
-For local use, applications such as Ollama, LMStudio, MLX-LM, llama.cpp, and KTransformers have also supported Qwen3.
+**Environment Setup**
 
-## Switching Between Thinking and Non-Thinking Mode
+| Item | Version          | 
+| ------------- | ------------------------------------------------------------ |  
+| Docker Version                  | Docker version 28.1.0, build 4d8c241 | 
+| Operating System                | Ubuntu 22.04.5 LTS    | 
+| FlagScale                       | Version: 0.8.0                        | 
+| FlagGems                        | Version: 3.0                          | 
 
-> [!TIP]
-> The `enable_thinking` switch is also available in APIs created by SGLang and vLLM. 
-> Please refer to our documentation for [SGLang](https://qwen.readthedocs.io/en/latest/deployment/sglang.html#thinking-non-thinking-modes) and [vLLM](https://qwen.readthedocs.io/en/latest/deployment/vllm.html#thinking-non-thinking-modes) users.
+## Operation Steps
 
-### `enable_thinking=True`
+### Download Open-source Model Weights
 
-By default, Qwen3 has thinking capabilities enabled, similar to QwQ-32B. This means the model will use its reasoning abilities to enhance the quality of generated responses. For example, when explicitly setting `enable_thinking=True` or leaving it as the default value in `tokenizer.apply_chat_template`, the model will engage its thinking mode.
+```bash
+pip install modelscope
+modelscope download --model Qwen/Qwen3-32B --local_dir /share/Qwen3-32B
 
-```python
-text = tokenizer.apply_chat_template(
-    messages,
-    tokenize=False,
-    add_generation_prompt=True,
-    enable_thinking=True  # True is the default value for enable_thinking
-)
 ```
 
-In this mode, the model will generate think content wrapped in a `<think>...</think>` block, followed by the final response.
-
-> [!NOTE]
-> For thinking mode, use `Temperature=0.6`, `TopP=0.95`, `TopK=20`, and `MinP=0` (the default setting in `generation_config.json`). **DO NOT use greedy decoding**, as it can lead to performance degradation and endless repetitions. For more detailed guidance, please refer to the [Best Practices](#best-practices) section.
+### Download FlagOS Image
 
-
-### `enable_thinking=False`
-
-We provide a hard switch to strictly disable the model's thinking behavior, aligning its functionality with the previous Qwen2.5-Instruct models. This mode is particularly useful in scenarios where disabling thinking is essential for enhancing efficiency.
-
-```python
-text = tokenizer.apply_chat_template(
-    messages,
-    tokenize=False,
-    add_generation_prompt=True,
-    enable_thinking=False  # Setting enable_thinking=False disables thinking mode
-)
+```bash
+docker pull harbor.baai.ac.cn/flagrelease-public/flagrelease_nvidia_qwen3sgl
 ```
 
-In this mode, the model will not generate any think content and will not include a `<think>...</think>` block.
-
-> [!NOTE]
-> For non-thinking mode, we suggest using `Temperature=0.7`, `TopP=0.8`, `TopK=20`, and `MinP=0`. For more detailed guidance, please refer to the [Best Practices](#best-practices) section.
-
-### Advanced Usage: Switching Between Thinking and Non-Thinking Modes via User Input
-
-We provide a soft switch mechanism that allows users to dynamically control the model's behavior when `enable_thinking=True`. Specifically, you can add `/think` and `/no_think` to user prompts or system messages to switch the model's thinking mode from turn to turn. The model will follow the most recent instruction in multi-turn conversations.
-
-Here is an example of a multi-turn conversation:
-
-```python
-from transformers import AutoModelForCausalLM, AutoTokenizer
+### Start the inference service
 
-class QwenChatbot:
-    def __init__(self, model_name="Qwen/Qwen3-32B"):
-        self.tokenizer = AutoTokenizer.from_pretrained(model_name)
-        self.model = AutoModelForCausalLM.from_pretrained(model_name)
-        self.history = []
-
-    def generate_response(self, user_input):
-        messages = self.history + [{"role": "user", "content": user_input}]
-
-        text = self.tokenizer.apply_chat_template(
-            messages,
-            tokenize=False,
-            add_generation_prompt=True
-        )
-
-        inputs = self.tokenizer(text, return_tensors="pt")
-        response_ids = self.model.generate(**inputs, max_new_tokens=32768)[0][len(inputs.input_ids[0]):].tolist()
-        response = self.tokenizer.decode(response_ids, skip_special_tokens=True)
-
-        # Update history
-        self.history.append({"role": "user", "content": user_input})
-        self.history.append({"role": "assistant", "content": response})
-
-        return response
-
-# Example Usage
-if __name__ == "__main__":
-    chatbot = QwenChatbot()
-
-    # First input (without /think or /no_think tags, thinking mode is enabled by default)
-    user_input_1 = "How many r's in strawberries?"
-    print(f"User: {user_input_1}")
-    response_1 = chatbot.generate_response(user_input_1)
-    print(f"Bot: {response_1}")
-    print("----------------------")
-
-    # Second input with /no_think
-    user_input_2 = "Then, how many r's in blueberries? /no_think"
-    print(f"User: {user_input_2}")
-    response_2 = chatbot.generate_response(user_input_2)
-    print(f"Bot: {response_2}") 
-    print("----------------------")
-
-    # Third input with /think
-    user_input_3 = "Really? /think"
-    print(f"User: {user_input_3}")
-    response_3 = chatbot.generate_response(user_input_3)
-    print(f"Bot: {response_3}")
+```bash
+#Container Startup
+docker run --rm --init --detach   --net=host --uts=host --ipc=host   --security-opt=seccomp=unconfined   --privileged=true   --ulimit stack=67108864   --ulimit memlock=-1   --ulimit nofile=1048576:1048576   --shm-size=32G   -v /share:/share   --gpus all   --name flagos   harbor.baai.ac.cn/flagrelease-public/flagrelease_nvidia_qwen3sgl   sleep infinity
 ```
 
-> [!NOTE]
-> For API compatibility, when `enable_thinking=True`, regardless of whether the user uses `/think` or `/no_think`, the model will always output a block wrapped in `<think>...</think>`. However, the content inside this block may be empty if thinking is disabled.
-> When `enable_thinking=False`, the soft switches are not valid. Regardless of any `/think` or `/no_think` tags input by the user, the model will not generate think content and will not include a `<think>...</think>` block.
-
-## Agentic Use
-
-Qwen3 excels in tool calling capabilities. We recommend using [Qwen-Agent](https://github.com/QwenLM/Qwen-Agent) to make the best use of agentic ability of Qwen3. Qwen-Agent encapsulates tool-calling templates and tool-calling parsers internally, greatly reducing coding complexity.
-
-To define the available tools, you can use the MCP configuration file, use the integrated tool of Qwen-Agent, or integrate other tools by yourself.
-```python
-from qwen_agent.agents import Assistant
-
-# Define LLM
-llm_cfg = {
-    'model': 'Qwen3-32B',
-
-    # Use the endpoint provided by Alibaba Model Studio:
-    # 'model_type': 'qwen_dashscope',
-    # 'api_key': os.getenv('DASHSCOPE_API_KEY'),
-
-    # Use a custom endpoint compatible with OpenAI API:
-    'model_server': 'http://localhost:8000/v1',  # api_base
-    'api_key': 'EMPTY',
-
-    # Other parameters:
-    # 'generate_cfg': {
-    #         # Add: When the response content is `<think>this is the thought</think>this is the answer;
-    #         # Do not add: When the response has been separated by reasoning_content and content.
-    #         'thought_in_content': True,
-    #     },
-}
-
-# Define Tools
-tools = [
-    {'mcpServers': {  # You can specify the MCP configuration file
-            'time': {
-                'command': 'uvx',
-                'args': ['mcp-server-time', '--local-timezone=Asia/Shanghai']
-            },
-            "fetch": {
-                "command": "uvx",
-                "args": ["mcp-server-fetch"]
-            }
-        }
-    },
-  'code_interpreter',  # Built-in tools
-]
+### Serve
 
-# Define Agent
-bot = Assistant(llm=llm_cfg, function_list=tools)
+```bash
+flagscale serve qwen3_next
 
-# Streaming generation
-messages = [{'role': 'user', 'content': 'https://qwenlm.github.io/blog/ Introduce the latest developments of Qwen'}]
-for responses in bot.run(messages=messages):
-    pass
-print(responses)
 ```
 
-## Processing Long Texts
-
-Qwen3 natively supports context lengths of up to 32,768 tokens. For conversations where the total length (including both input and output) significantly exceeds this limit, we recommend using RoPE scaling techniques to handle long texts effectively. We have validated the model's performance on context lengths of up to 131,072 tokens using the [YaRN](https://arxiv.org/abs/2309.00071) method.
-
-YaRN is currently supported by several inference frameworks, e.g., `transformers` and `llama.cpp` for local use, `vllm` and `sglang` for deployment. In general, there are two approaches to enabling YaRN for supported frameworks:
 
-- Modifying the model files:
-  In the `config.json` file, add the `rope_scaling` fields:
-    ```json
-    {
-        ...,
-        "rope_scaling": {
-            "rope_type": "yarn",
-            "factor": 4.0,
-            "original_max_position_embeddings": 32768
-        }
-    }
-    ```
-  For `llama.cpp`, you need to regenerate the GGUF file after the modification.
+## Service Invocation
 
-- Passing command line arguments:
+### API-based Invocation Script
 
-  For `vllm`, you can use
-    ```shell
-    vllm serve ... --rope-scaling '{"rope_type":"yarn","factor":4.0,"original_max_position_embeddings":32768}' --max-model-len 131072  
-    ```
-
-  For `sglang`, you can use
-    ```shell
-    python -m sglang.launch_server ... --json-model-override-args '{"rope_scaling":{"rope_type":"yarn","factor":4.0,"original_max_position_embeddings":32768}}'
-    ```
+```bash
+import openai
+openai.api_key = "EMPTY"
+openai.base_url = "http://<server_ip>:9010/v1/"
+model = "Qwen3-32B-nvidia-flagos"
+messages = [
+    {"role": "system", "content": "You are a helpful assistant."},
+    {"role": "user", "content": "What's the weather like today?"}
+]
+response = openai.chat.completions.create(
+    model=model,
+    messages=messages,
+    stream=False,
+)
+for item in response:
+    print(item)
 
-  For `llama-server` from `llama.cpp`, you can use
-    ```shell
-    llama-server ... --rope-scaling yarn --rope-scale 4 --yarn-orig-ctx 32768
-    ```
+```
 
-> [!IMPORTANT]
-> If you encounter the following warning
-> ```
-> Unrecognized keys in `rope_scaling` for 'rope_type'='yarn': {'original_max_position_embeddings'}
-> ```
-> please upgrade `transformers>=4.51.0`.
+### AnythingLLM Integration Guide
 
-> [!NOTE]
-> All the notable open-source frameworks implement static YaRN, which means the scaling factor remains constant regardless of input length, **potentially impacting performance on shorter texts.**
-> We advise adding the `rope_scaling` configuration only when processing long contexts is required. 
-> It is also recommended to modify the `factor` as needed. For example, if the typical context length for your application is 65,536 tokens, it would be better to set `factor` as 2.0. 
+#### 1. Download & Install
 
-> [!NOTE]
-> The default `max_position_embeddings` in `config.json` is set to 40,960. This allocation includes reserving 32,768 tokens for outputs and 8,192 tokens for typical prompts, which is sufficient for most scenarios involving short text processing. If the average context length does not exceed 32,768 tokens, we do not recommend enabling YaRN in this scenario, as it may potentially degrade model performance.
+- Visit the official site: https://anythingllm.com/
+- Choose the appropriate version for your OS (Windows/macOS/Linux)
+- Follow the installation wizard to complete the setup
 
-> [!TIP]
-> The endpoint provided by Alibaba Model Studio supports dynamic YaRN by default and no extra configuration is needed.
+#### 2. Configuration
 
-## Best Practices
+- Launch AnythingLLM
+- Open settings (bottom left, fourth tab)
+- Configure core LLM parameters
+- Click "Save Settings" to apply changes
 
-To achieve optimal performance, we recommend the following settings:
+#### 3. Model Interaction
 
-1. **Sampling Parameters**:
-   - For thinking mode (`enable_thinking=True`), use `Temperature=0.6`, `TopP=0.95`, `TopK=20`, and `MinP=0`. **DO NOT use greedy decoding**, as it can lead to performance degradation and endless repetitions.
-   - For non-thinking mode (`enable_thinking=False`), we suggest using `Temperature=0.7`, `TopP=0.8`, `TopK=20`, and `MinP=0`.
-   - For supported frameworks, you can adjust the `presence_penalty` parameter between 0 and 2 to reduce endless repetitions. However, using a higher value may occasionally result in language mixing and a slight decrease in model performance.
+- After model loading is complete:
+  - Click **"New Conversation"**
+  - Enter your question (e.g., “Explain the basics of quantum computing”)
+  - Click the send button to get a response
 
-2. **Adequate Output Length**: We recommend using an output length of 32,768 tokens for most queries. For benchmarking on highly complex problems, such as those found in math and programming competitions, we suggest setting the max output length to 38,912 tokens. This provides the model with sufficient space to generate detailed and comprehensive responses, thereby enhancing its overall performance.
+# Contributing
 
-3. **Standardize Output Format**: We recommend using prompts to standardize model outputs when benchmarking.
-   - **Math Problems**: Include "Please reason step by step, and put your final answer within \boxed{}." in the prompt.
-   - **Multiple-Choice Questions**: Add the following JSON structure to the prompt to standardize responses: "Please show your choice in the `answer` field with only the choice letter, e.g., `"answer": "C"`."
+We warmly welcome global developers to join us:
 
-4. **No Thinking Content in History**: In multi-turn conversations, the historical model output should only include the final output part and does not need to include the thinking content. It is implemented in the provided chat template in Jinja2. However, for frameworks that do not directly use the Jinja2 chat template, it is up to the developers to ensure that the best practice is followed.
+1. Submit Issues to report problems
+2. Create Pull Requests to contribute code
+3. Improve technical documentation
+4. Expand hardware adaptation support
 
-### Citation
 
-If you find our work helpful, feel free to give us a cite.
+# License
 
-```
-@misc{qwen3technicalreport,
-      title={Qwen3 Technical Report}, 
-      author={Qwen Team},
-      year={2025},
-      eprint={2505.09388},
-      archivePrefix={arXiv},
-      primaryClass={cs.CL},
-      url={https://arxiv.org/abs/2505.09388}, 
-}
-```
+本模型的权重来源于Qwen/Qwen3-32B，以apache2.0协议https://www.apache.org/licenses/LICENSE-2.0.txt开源。