minimax-m2.1-reap-observations

> [!TIP] > Support this work: **[donate.sybilsolutions.ai](https://donate.sybilsolutions.ai)** > > REAP surfaces: [GLM](https://huggingface.co/spaces/0xSero/reap-glm-family) | [MiniMax](https://huggingface.co/spaces/0xSero/reap-minimax-family) | [Qwen](https://huggingface.co/spaces/0xSero/reap-qwen-family) | [Gemma](https://huggingface.co/spaces/0xSero/reap-gemma-family) | [Paper](https://arxiv.org/abs/2510.13999) | [Code](https://github.com/CerebrasResearch/reap) | [PR17](https://github.com/CerebrasResearch/reap/pull/17) | [Cerebras Collection](https://huggingface.co/collections/cerebras/cerebras-reap) # MiniMax-M2.1 REAP Stress Test Observations Comprehensive stress test results for MiniMax-M2.1 models pruned with REAP (Router-weighted Expert Activation Pruning) at various compression ratios. ## Dataset Description This dataset contains 96 stress test results across 4 pruned MiniMax-M2.1 models, testing for repetition loops at various temperatures. ### Models Tested | Model | Compression | Experts | Status | |-------|-------------|---------|--------| | REAP-20 | 20% | 204/256 | Deprecated (1 loop) | | REAP-30 | 30% | 180/256 | **Recommended** (0 loops) | | REAP-40 | 40% | 154/256 | **Recommended** (0 loops) | | REAP-50 | 50% | 128/256 | Deprecated (2 loops) | ### Test Configuration - **Temperatures**: 0.0, 0.2, 0.7, 1.0 - **Prompt Types**: loop_count, code_fib, math_word, tool_json, essay, repeat_abc - **Max Tokens**: 512 - **Loop Detection**: 5+ repeated lines ## Key Findings ### The "Goldilocks Zone" Only **30% and 40% pruning** produce stable models: - 20% pruning: Loops on math_word at temp 0.0 - 30% pruning: **0 loops** (recommended) - 40% pruning: **0 loops** (recommended) - 50% pruning: Loops on math_word at temps 0.0 and 0.2 ### Temperature Sensitivity Loops only occur at low temperatures (0.0-0.2). Recommended minimum: **0.3+** ## Dataset Schema ```json { "model": "MiniMax-M2.1-REAP-50p", "temp": 0.0, "prompt": "math_word", "input_tokens": 64, "output_tokens": 512, "elapsed_sec": 102.42, "tok_per_sec": 5.0, "is_loop": true, "completion_preview": "...", "completion_end": "..." } ``` ## Recommended Models - [MiniMax-M2.1-REAP-30](https://huggingface.co/0xSero/MiniMax-M2.1-REAP-30) - 162B params - [MiniMax-M2.1-REAP-40](https://huggingface.co/0xSero/MiniMax-M2.1-REAP-40) - 139B params ## Acknowledgments - Compute sponsored by [Prime Intellect](https://www.primeintellect.ai/) - REAP implementation by [Cerebras](https://github.com/Cerebras/reap) - Base model by [MiniMax](https://huggingface.co/MiniMaxAI/MiniMax-M2.1)