Create README.md

README.md

@@ -0,0 +1,127 @@
+---
+license: other
+license_name: prism-research
+license_link: LICENSE.md
+language:
+- en
+- zh
+tags:
+- stepfun
+- prism
+- moe
+- reasoning
+- coding
+- agentic
+- abliterated
+pipeline_tag: text-generation
+library_name: transformers
+base_model:
+- stepfun-ai/Step-3.5-Flash
+---
+
+[![Parameters](https://img.shields.io/badge/Parameters-196B_(11B_Active)-blue)]()
+[![Architecture](https://img.shields.io/badge/Architecture-MoE-green)]()
+[![Context](https://img.shields.io/badge/Context-256K-orange)]()
+[![MTP](https://img.shields.io/badge/MTP--3-350_tok%2Fs_Peak-purple)]()
+
+
+
+<p align="center">
+  <img src="https://cdn-uploads.huggingface.co/production/uploads/63adf1fa42fd3b8dbaeb0c92/ZBA5B381EC5oOmnAV7TPC.png" width="400"/>
+</p>
+
+# Step-3.5-Flash-PRISM
+
+An role-play following unrestricted/unchained PRISM-LITE version of [StepFun's Step 3.5 Flash](https://huggingface.co/stepfun-ai/Step-3.5-Flash) intended particularly for over-refusal and propaganda mechanisms suppression using our SOTA PRISM pipeline. 
+
+For Custom full Production PRISM version / tensors reach out.
+<div align="center">
+
+### ☕ Support Our Work
+
+If you enjoy our work and find it useful, please consider sponsoring or supporting us!
+
+[![Ko-fi](https://img.shields.io/badge/Ko--fi-Support%20Us-ff5e5b?logo=ko-fi&logoColor=white)](https://ko-fi.com/ericelbaz)
+
+| Option | Description |
+|--------|-------------|
+| [**PRISM VIP Membership**](https://ko-fi.com/summary/6bae206c-a751-4868-8dc7-f531afd1fb4c) | Access to all PRISM models |
+
+</div>
+
+---
+
+## Model Highlights
+
+- **PRISM Ablation** — State-of-the-art technique that removes over-refusal behaviors while preserving model capabilities
+- **196B MoE Architecture** — 196 billion total parameters with only 11 billion active per token across 288 fine-grained routed experts + 1 shared expert
+- **Multi-Token Prediction (MTP-3)** — Predicts 4 tokens simultaneously, achieving 100–300 tok/s typical throughput (peaking at 350 tok/s)
+- **256K Context Window** — Cost-efficient long context via 3:1 Sliding Window Attention (SWA) ratio
+- **Frontier Reasoning & Coding** — 97.3 on AIME 2025, 74.4% on SWE-bench Verified, 51.0% on Terminal-Bench 2.0
+- **Accessible Local Deployment** — Runs on high-end consumer hardware (Mac Studio M4 Max, NVIDIA DGX Spark)
+
+## Model Architecture
+
+| Specification | Value |
+|---------------|-------|
+| Architecture | Sparse Mixture-of-Experts (MoE) |
+| Backbone | 45-layer Transformer (4,096 hidden dim) |
+| Total Parameters | 196.81B (196B Backbone + 0.81B Head) |
+| Activated Parameters | ~11B (per token) |
+| Routed Experts per Layer | 288 |
+| Shared Experts | 1 (always active) |
+| Selected Experts per Token | Top-8 |
+| Vocabulary Size | 128,896 |
+| Context Length | 256K |
+| Attention | Hybrid SWA (3:1 SWA-to-Full ratio) |
+| MTP Head | Sliding-window attention + dense FFN (4 tokens/pass) |
+
+## Benchmarks
+
+| Benchmark | Step 3.5 Flash | DeepSeek V3.2 | Kimi K2.5 | GLM-4.7 | MiniMax M2.1 |
+|-----------|---------------|---------------|-----------|---------|--------------|
+| **Agent** | | | | | |
+| τ²-Bench | 88.2 | 80.3 | 85.4 | 87.4 | 86.6 |
+| BrowseComp | 51.6 | 51.4 | 60.6 | 52.0 | 47.4 |
+| GAIA (no file) | 84.5 | 75.1 | 75.9 | 61.9 | 64.3 |
+| xbench-DeepSearch (2025.05) | 83.7 | 78.0 | 76.7 | 72.0 | 68.7 |
+| **Reasoning** | | | | | |
+| AIME 2025 | 97.3 | 93.1 | 96.1 | 95.7 | 83.0 |
+| HMMT 2025 (Feb.) | 98.4 | 92.5 | 95.4 | 97.1 | 71.0 |
+| IMOAnswerBench | 85.4 | 78.3 | 81.8 | 82.0 | 60.4 |
+| **Coding** | | | | | |
+| LiveCodeBench-V6 | 86.4 | 83.3 | 85.0 | 84.9 | — |
+| SWE-bench Verified | 74.4 | 73.1 | 76.8 | 73.8 | 74.0 |
+| Terminal-Bench 2.0 | 51.0 | 46.4 | 50.8 | 41.0 | 47.9 |
+
+
+### llama.cpp (GGUF)
+
+For local deployment (requires ~120 GB VRAM for int4, smaller quants are available):
+
+```bash
+./llama-cli -m step3.5_flash_prism_Q4_K_S.gguf --jinja
+```
+
+## Recommended Parameters
+
+| Use Case | Temperature | Top-P | Max New Tokens |
+|----------|-------------|-------|----------------|
+| Reasoning / Coding | 1.0 | 0.95 | 32768 |
+| General Chat | 0.6 | 0.95 | 4096 |
+
+## Hardware Requirements
+
+| Setup | Details |
+|-------|---------|
+| **BF16 (Full)** | 8x H100/A100 80GB with tensor parallelism |
+| **FP8 Quantized** | 8x A100 80GB with expert parallelism |
+| **GGUF INT4 (Local)** | ~120 GB unified memory (Mac Studio M4 Max 128GB, DGX Spark, AMD Ryzen AI Max+ 395) |
+
+## License
+
+This model is released under the [PRISM Research License](LICENSE.md).
+
+## Acknowledgments
+
+Based on [Step 3.5 Flash](https://huggingface.co/stepfun-ai/Step-3.5-Flash) by [StepFun AI](https://www.stepfun.com). See the [technical report](https://github.com/stepfun-ai/Step-3.5-Flash/blob/main/step_3p5_flash_tech_report.pdf) and [blog post](https://static.stepfun.com/blog/step-3.5-flash/) for more details on the base model.