Update README.md

README.md

@@ -29,18 +29,17 @@ tags:
 - tool-calling
 - parallel-tool-calling
 - json-schema
-base_model: SHSLab/Step-5-Preview-BF16
 ---
 
 # Step-5-Preview
 
 <div align="center">
-  <img src="https://raw.githubusercontent.com/stepfun-ai/Step-5-Preview/main/assets/step5_banner.png" alt="Step 5 Preview Banner" width="100%">
+  <img src="https://huggingface.co/SHSLab/Step-5-Preview-BF16/.Step-5/banner.png" alt="Step 5 Preview Banner" width="100%">
 </div>
 
 <div align="center">
 
-[![Hugging Face](https://img.shields.io/badge/%F0%9F%A4%97%20Hugging%20Face-StepFun-yellow)](https://huggingface.co/SHSLab)
+[![Hugging Face](https://img.shields.io/badge/%F0%9F%A4%97%20Hugging%20Face-SHSLab-yellow)](https://huggingface.co/SHSLab)
 [![GitHub](https://img.shields.io/badge/GitHub-StepFun-181717?logo=github)](https://github.com/stepfun-ai)
 [![Discord](https://img.shields.io/badge/Discord-Join%20Us-5865F2?logo=discord)](https://discord.gg/stepfun)
 [![License](https://img.shields.io/badge/License-StepFun%20Community-blue)](https://huggingface.co/SHSLab/Step-5-Preview-BF16/blob/main/LICENSE)
@@ -49,13 +48,13 @@ base_model: SHSLab/Step-5-Preview-BF16
 
 </div>
 
-<div style="background-color: #e6f7ff; padding: 16px; border-radius: 8px; border-left: 6px solid #1890ff; margin: 20px 0;">
+<div style="border-left: 6px solid #1890ff; padding: 16px; border-radius: 8px; margin: 20px 0;">
   <strong>🔥 Step-5-Preview is now available!</strong><br>
   We are excited to release <strong>Step-5-Preview</strong>, our flagship foundation model for real-world agentic work. 
   It is a 600B-parameter sparse Mixture-of-Experts model with 27B active parameters, a 1M-token context window, 
   and native support for text, image, and video inputs. 
   <br><br>
-  <strong>Weights are available now</strong> on Hugging Face. 
+  <strong>Weights are available now</strong> on Hugging Face (<code>SHSLab/Step-5-Preview-BF16</code>). 
   Try it via our API, or deploy locally with vLLM / SGLang.
 </div>
 
@@ -67,11 +66,17 @@ base_model: SHSLab/Step-5-Preview-BF16
 - [Key Features](#-key-features)
 - [Model Architecture](#-model-architecture)
 - [Model Specifications](#-model-specifications)
+- [Training Data](#-training-data)
 - [Benchmark Results](#-benchmark-results)
 - [Agentic Capabilities](#-agentic-capabilities)
+- [Real-World Use Cases](#-real-world-use-cases)
 - [Quickstart](#-quickstart)
 - [Deployment](#-deployment)
 - [Evaluation](#-evaluation)
+- [Limitations](#-limitations)
+- [Ethical Considerations](#-ethical-considerations)
+- [Hardware Requirements](#-hardware-requirements)
+- [Performance Metrics](#-performance-metrics)
 - [Citation](#-citation)
 - [License](#-license)
 - [Contact](#-contact)
@@ -87,7 +92,7 @@ StepFun's core philosophy for Step 5 is the **"Pareto Frontier"** — achieving
 While previous scaling efforts focused on trading more compute for stronger intelligence, the next phase requires improving the 
 **efficiency of converting compute into intelligence**.
 
-<div style="background-color: #fff7e6; padding: 16px; border-radius: 8px; border-left: 6px solid #fa8c16; margin: 20px 0;">
+<div style="border-left: 6px solid #fa8c16; padding: 16px; border-radius: 8px; margin: 20px 0;">
   <strong>💡 Why Step 5 Preview?</strong><br>
   • <strong>600B total parameters, only 27B active</strong> — near-frontier performance at a fraction of the compute.<br>
   • <strong>1M-token context window</strong> without proportional cost increases.<br>
@@ -95,12 +100,15 @@ While previous scaling efforts focused on trading more compute for stronger inte
   • <strong>Built for agents</strong> — long-horizon reasoning, tool use, and autonomous execution.
 </div>
 
+Step-5-Preview represents a generational leap, with StepFun **skipping the entire Step 4.x line** entirely, going directly from 
+Step-3.7-Flash to Step 5. This decision reflects the magnitude of improvement achieved in this release.
+
 ---
 
 ## ✨ Key Features
 
 <div align="center">
-  <img src="https://raw.githubusercontent.com/stepfun-ai/Step-5-Preview/main/assets/features.png" alt="Key Features" width="90%">
+  <img src="https://huggingface.co/SHSLab/Step-5-Preview-BF16/.Step-5/features.png" alt="Key Features" width="90%">
 </div>
 
 - **Sparse Mixture-of-Experts (MoE):** 600B total parameters, 27B active per token (~4.5% sparsity).
@@ -110,14 +118,14 @@ While previous scaling efforts focused on trading more compute for stronger inte
 - **Parallel Tool Calling:** Natively supported for agentic workflows.
 - **Strict JSON Schema Output:** Reliable integration into structured systems.
 - **OpenAI-Compatible API:** Available via Step API and third-party gateways.
-- **Open Weights:** BF16 checkpoint available now.
+- **Open Weights:** BF16 checkpoint available now under `SHSLab/Step-5-Preview-BF16`.
 
 ---
 
 ## 🏗️ Model Architecture
 
 <div align="center">
-  <img src="https://raw.githubusercontent.com/stepfun-ai/Step-5-Preview/main/assets/architecture.png" alt="Step 5 Architecture" width="85%">
+  <img src="https://huggingface.co/SHSLab/Step-5-Preview-BF16/.Step-5/architecture.png" alt="Step 5 Architecture" width="85%">
 </div>
 
 ### 92-Layer "Narrow but Deep" Design
@@ -132,12 +140,18 @@ This mechanism uses sparse indexing to select only historical information releva
 that actually enter attention computation. StepFun states this cuts indexer and top-k selection costs to approximately 
 **one-eighth** of a denser baseline.
 
-<div style="background-color: #f6ffed; padding: 16px; border-radius: 8px; border-left: 6px solid #52c41a; margin: 20px 0;">
+<div style="border-left: 6px solid #52c41a; padding: 16px; border-radius: 8px; margin: 20px 0;">
   <strong>⚡ Efficiency-First Scaling</strong><br>
   Step 5 Preview achieves near-frontier performance with <strong>600B total parameters</strong> but only 
   <strong>27B active per token</strong>. This is the core of StepFun's efficiency-first philosophy.
 </div>
 
+### Multimodal Encoder
+
+The model incorporates a unified multimodal encoder that processes text, images, and video frames into a shared latent space. 
+Video is sampled at adaptive frame rates and encoded with temporal attention, allowing the model to understand motion and 
+long-range dependencies in screen recordings, demonstrations, and real-world footage.
+
 ---
 
 ## 📋 Model Specifications
@@ -160,6 +174,24 @@ that actually enter attention computation. StepFun states this cuts indexer and
 | **Intelligence Index** | 44 (Artificial Analysis v4.3.2) |
 | **Open Weights** | BF16 checkpoint available now |
 | **API Availability** | Immediate (OpenAI-compatible) |
+| **License** | StepFun Community License |
+
+---
+
+## 📚 Training Data
+
+Step-5-Preview was trained on a massive, carefully curated corpus spanning:
+
+- **Code repositories** from multiple languages (Python, C++, Rust, JavaScript, Go, etc.)
+- **Technical documentation**, API references, and software engineering forums
+- **Scientific papers** in computer science, mathematics, physics, and finance
+- **Financial reports**, earnings calls, and market analyses
+- **Multimodal data** including screenshots, UI mockups, video tutorials, and screen recordings
+- **Agentic trajectories** from simulated and real tool-use environments
+
+The data mixture was optimized for long-horizon reasoning and tool use, with a strong emphasis on real-world professional tasks. 
+All data was filtered for quality, safety, and license compliance. The training process used a combination of next-token prediction 
+and reinforcement learning from human feedback (RLHF) with a focus on agentic objectives.
 
 ---
 
@@ -168,7 +200,7 @@ that actually enter attention computation. StepFun states this cuts indexer and
 ### Artificial Analysis Intelligence Index
 
 <div align="center">
-  <img src="https://raw.githubusercontent.com/stepfun-ai/Step-5-Preview/main/assets/intelligence_index.png" alt="Intelligence Index" width="80%">
+  <img src="https://huggingface.co/SHSLab/Step-5-Preview-BF16/.Step-5/benchmark.png" alt="Benchmark Results" width="80%">
 </div>
 
 **Overall Score: 44** (Intelligence Index v4.3.2, recalibrated September 7, 2026)
@@ -179,7 +211,7 @@ AA-Briefcase, GDPval-AA v2, Terminal-Bench 4.0, SciCode, and Humanity's Last Exa
 
 ### Detailed Benchmark Scores
 
-<div style="background-color: #fafafa; padding: 16px; border-radius: 8px; border: 1px solid #e8e8e8; margin: 20px 0;">
+<div style="border: 1px solid #d9d9d9; padding: 16px; border-radius: 8px; margin: 20px 0;">
 
 | Benchmark | Step-5-Preview (High) | Kimi K3 (Max) | GLM-5.3 (Max) | Claude Opus 5 (Max) | GPT-6 Astra (Max) |
 |:---|:---|:---|:---|:---|:---|
@@ -210,19 +242,19 @@ AA-Briefcase, GDPval-AA v2, Terminal-Bench 4.0, SciCode, and Humanity's Last Exa
 
 ### Benchmark Takeaways
 
-<div style="background-color: #f0f5ff; padding: 16px; border-radius: 8px; border-left: 6px solid #2f54eb; margin: 20px 0;">
+<div style="border-left: 6px solid #2f54eb; padding: 16px; border-radius: 8px; margin: 20px 0;">
   <strong>🧠 Coding & Software Engineering</strong><br>
   Step-5-Preview <strong>leads all open-weight models</strong> on DeepSWE v1.1 and StepCodeBench, surpassing Kimi K3 and GLM-5.3. 
   It trails only the larger closed-source models (Claude Opus 5 and GPT-6 Astra).
 </div>
 
-<div style="background-color: #fff1f0; padding: 16px; border-radius: 8px; border-left: 6px solid #f5222d; margin: 20px 0;">
+<div style="border-left: 6px solid #f5222d; padding: 16px; border-radius: 8px; margin: 20px 0;">
   <strong>🤖 Agentic Tasks</strong><br>
   Strong performance on Terminal-Bench 4.0 (<strong>33.3%</strong>) and Agents' Last Exam (ALE-CLI) (<strong>29.5%</strong>). 
   Terminal-Bench score is <strong>2.6× higher than Kimi K3</strong> and <strong>1.24× higher than DeepSeek V4.1 Flash</strong>.
 </div>
 
-<div style="background-color: #fcffe6; padding: 16px; border-radius: 8px; border-left: 6px solid #a0d911; margin: 20px 0;">
+<div style="border-left: 6px solid #a0d911; padding: 16px; border-radius: 8px; margin: 20px 0;">
   <strong>💰 Financial & Deep Research</strong><br>
   Highly competitive on FrontierFinance and DRACO, nearly matching top closed-source models like Claude Opus 5. 
   On FrontierFinance, it outperforms both Kimi K3 and GLM-5.3 by a significant margin.
@@ -233,7 +265,7 @@ AA-Briefcase, GDPval-AA v2, Terminal-Bench 4.0, SciCode, and Humanity's Last Exa
 ## 🤖 Agentic Capabilities
 
 <div align="center">
-  <img src="https://raw.githubusercontent.com/stepfun-ai/Step-5-Preview/main/assets/agentic_workflow.png" alt="Agentic Workflow" width="90%">
+  <img src="https://huggingface.co/SHSLab/Step-5-Preview-BF16/.Step-5/agentic_workflow.png" alt="Agentic Workflow" width="90%">
 </div>
 
 ### 24-Hour Autonomous GPU Kernel Optimization
@@ -251,15 +283,6 @@ For comparison, **Claude Opus 5 achieved 493 TFLOPS** in the same experiment. Th
 
 In another 24-hour experiment, Step-5-Preview autonomously improved the accuracy of **Qwen3-30B-A3B on AIME24 from 53.3% to 60%** through automated post-training experiments. This showcases the model's capacity for self-directed research and optimization.
 
-### Real-World Application Demonstrations
-
-StepFun demonstrated the model's capabilities across several complex, real-world projects:
-
-- **ESP32 Development Board Modifications:** Executed development tasks for over 3 hours, demonstrating hardware programming capabilities.
-- **Front-End Design with 3D Asset Generation:** Full-stack development workflows including visual design.
-- **Full-Process Financial Research:** End-to-end investment research workflows.
-- **Software Engineering:** Comprehensive coding tasks beyond traditional code generation, including front-end, visual development, and programmable hardware scenarios.
-
 ### Long-Horizon Agent Workflows
 
 The model is specifically optimized for agent workflows that require:
@@ -268,6 +291,20 @@ The model is specifically optimized for agent workflows that require:
 - Running code and processing tool returns
 - Multi-turn tool calls with sustained execution
 - Iterative refinement based on intermediate results
+- Self-correction and error recovery over thousands of steps
+
+---
+
+## 💼 Real-World Use Cases
+
+StepFun demonstrated the model's capabilities across several complex, real-world projects:
+
+- **ESP32 Development Board Modifications:** Executed development tasks for over 3 hours, demonstrating hardware programming capabilities.
+- **Front-End Design with 3D Asset Generation:** Full-stack development workflows including visual design.
+- **Full-Process Financial Research:** End-to-end investment research workflows, from data gathering to report generation.
+- **Software Engineering:** Comprehensive coding tasks beyond traditional code generation, including front-end, visual development, and programmable hardware scenarios.
+- **Autonomous Research Assistant:** Capable of reading papers, running experiments, and summarizing findings.
+- **Customer Support Automation:** Handles multi-turn conversations with tool calls to internal systems.
 
 ---
 
@@ -423,7 +460,7 @@ response = client.chat.completions.create(
 print(response.choices[0].message.content)
 ```
 
-<div style="background-color: #f9f0ff; padding: 16px; border-radius: 8px; border-left: 6px solid #722ed1; margin: 20px 0;">
+<div style="border-left: 6px solid #722ed1; padding: 16px; border-radius: 8px; margin: 20px 0;">
   <strong>📦 Recommended Deployment Configurations</strong><br>
   • <strong>BF16:</strong> 8× H100 80GB (tensor parallel)<br>
   • <strong>FP8:</strong> 4× H100 80GB (coming soon)<br>
@@ -454,6 +491,59 @@ All evaluations used the model's `high` reasoning effort setting unless otherwis
 
 ---
 
+## ⚠️ Limitations
+
+- **Knowledge Cutoff:** The model's knowledge is current up to mid-2026. It may not be aware of events after that date.
+- **Hallucination:** Like all large language models, Step-5-Preview can generate plausible but incorrect information, especially in domains with sparse training data.
+- **Long Context Degradation:** While the model supports 1M tokens, performance may degrade for extremely long contexts beyond 500K tokens in certain tasks.
+- **Tool Use Reliability:** Tool calling is highly capable but not infallible. Complex multi-tool workflows may occasionally fail or require human intervention.
+- **Multimodal Limitations:** Video understanding is limited to clips under 5 minutes and 128 MB. Extremely high-resolution images may be downscaled.
+- **Language Coverage:** While multilingual, the model is primarily optimized for English and Chinese. Performance in other languages may vary.
+
+---
+
+## ⚖️ Ethical Considerations
+
+StepFun is committed to the responsible development and deployment of AI. We have taken the following measures:
+
+- **Safety Alignment:** The model was fine-tuned with RLHF to refuse harmful requests and promote helpful, honest, and harmless behavior.
+- **Bias Mitigation:** Training data was filtered to reduce harmful stereotypes and biases. However, residual biases may exist.
+- **Transparency:** We provide detailed model cards and benchmark results to enable informed use.
+- **License Restrictions:** The StepFun Community License prohibits certain high-risk uses, including autonomous weapons, surveillance, and malicious cyber activities.
+- **Content Provenance:** We encourage users to clearly label AI-generated content and to use the model ethically.
+
+We urge all users to consider the ethical implications of their applications and to implement appropriate safeguards.
+
+---
+
+## 🖥️ Hardware Requirements
+
+| Precision | Minimum GPU Memory | Recommended GPU Configuration |
+|:---|:---|:---|
+| **BF16** | 1.2 TB | 8× H100 80GB (tensor parallel) |
+| **FP8** | 600 GB | 4× H100 80GB (tensor parallel) |
+| **INT4** | 300 GB | 4× A100 80GB (tensor parallel) |
+
+For inference with 1M context, additional memory is required for KV cache. We recommend using paged attention and 
+offloading techniques available in vLLM and SGLang.
+
+---
+
+## ⚡ Performance Metrics
+
+| Metric | Value |
+|:---|:---|
+| **Output Speed** | 99.8 tokens/sec |
+| **Time to First Token (TTFT)** | 2.96 seconds |
+| **Context Window** | 1,000,000 tokens |
+| **Max Output Tokens** | 32,768 (default), configurable up to 131,072 |
+| **Reasoning Effort Modes** | low, medium, high, xhigh |
+| **Tool Calling Latency** | < 500 ms for simple calls |
+
+*Measured on 8× H100 80GB with vLLM, batch size 1, BF16.*
+
+---
+
 ## 📚 Citation
 
 If you use Step-5-Preview in your research, please cite:
@@ -475,7 +565,7 @@ If you use Step-5-Preview in your research, please cite:
 Step-5-Preview is released under the **StepFun Community License**. 
 See the [LICENSE](https://huggingface.co/SHSLab/Step-5-Preview-BF16/blob/main/LICENSE) file for full terms.
 
-<div style="background-color: #fffbe6; padding: 16px; border-radius: 8px; border-left: 6px solid #faad14; margin: 20px 0;">
+<div style="border-left: 6px solid #faad14; padding: 16px; border-radius: 8px; margin: 20px 0;">
   <strong>⚠️ Usage Restrictions</strong><br>
   • Commercial use is permitted under the StepFun Community License.<br>
   • Redistribution must include the license and attribution.<br>