[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill:01M6NB3A70FAJW8FJS5M8MF78S":3,"skill-same:01M6NB3A70FAJW8FJS5M8MF78S":87},{"skillNo":4,"name":5,"nameZh":6,"description":7,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":16,"categories":17,"keywords":23,"publishedAt":44,"linkOnly":45,"markdownContent":46,"markdownBody":47,"frontMatter":48,"refName":56,"commitSha":57,"skillDirectory":58,"forksCount":14,"repositoryHtmlUrl":59,"repositoryHomepageUrl":6,"ownerHtmlUrl":60,"ownerBlogUrl":6,"ownerDescription":61,"updatedAt":62,"install":63,"resources":76},"01M6NB3A70FAJW8FJS5M8MF78S","awq-quantization",null,"Activation-aware weight quantization for 4-bit LLM compression with 3x speedup and minimal accuracy loss. Use when deploying large models (7B-70B) on limited GPU memory, when you need faster inference than GPTQ with better accuracy preservation, or for instruction-tuned and multimodal models. MLSys 2024 Best Paper Award winner.","zhengliangliahu-cloud\u002Fskills","zhengliangliahu-cloud","https:\u002F\u002Favatars.githubusercontent.com\u002Fu\u002F248268067?v=4","01M6NAQ109P2Y11XSV5D9QJFFY","Lin","01M6NAQ15R1M4E18CA2KPJHC06",1,"MIT","https:\u002F\u002Fgithub.com\u002Fzhengliangliahu-cloud\u002Fskills\u002Fblob\u002Fmain\u002FAI-Research-SKILLs\u002F10-optimization\u002Fawq\u002FSKILL.md",[18],{"categoryNo":19,"categoryName":20,"taxonomyType":21,"taxonomyTypeDesc":22},"01JBSEEDCAT000000000000021","资料检索与研究","SCENARIO","使用场景",[24,25,26,27,28,29,30,31,32,33,34,35,36,37,38,39,40,41,42,43],"awq quantization","Quantization","AWQ","Activation-Aware","accuracy","for","orchestra","you","GPTQ","Marlin Kernels","vLLM Integration","Memory Optimization","memory","4-Bit","bit","activation","Transformers","Fast Inference","Optimization","multimodal","2026-11-10 10:12:30",false,"---\nname: awq-quantization\ndescription: Activation-aware weight quantization for 4-bit LLM compression with 3x speedup and minimal accuracy loss. Use when deploying large models (7B-70B) on limited GPU memory, when you need faster inference than GPTQ with better accuracy preservation, or for instruction-tuned and multimodal models. MLSys 2024 Best Paper Award winner.\nversion: 1.0.0\nauthor: Orchestra Research\nlicense: MIT\ntags: [Optimization, AWQ, Quantization, 4-Bit, Activation-Aware, Memory Optimization, Fast Inference, vLLM Integration, Marlin Kernels]\ndependencies: [autoawq, transformers>=4.45.0, torch>=2.0.0]\n---\n\n# AWQ (Activation-aware Weight Quantization)\n\n4-bit quantization that preserves salient weights based on activation patterns, achieving 3x speedup with minimal accuracy loss.\n\n## When to use AWQ\n\n**Use AWQ when:**\n- Need 4-bit quantization with \u003C5% accuracy loss\n- Deploying instruction-tuned or chat models (AWQ generalizes better)\n- Want ~2.5-3x inference speedup over FP16\n- Using vLLM for production serving\n- Have Ampere+ GPUs (A100, H100, RTX 40xx) for Marlin kernel support\n\n**Use GPTQ instead when:**\n- Need maximum ecosystem compatibility (more tools support GPTQ)\n- Working with ExLlamaV2 backend specifically\n- Have older GPUs without Marlin support\n\n**Use bitsandbytes instead when:**\n- Need zero calibration overhead (quantize on-the-fly)\n- Want to fine-tune with QLoRA\n- Prefer simpler integration\n\n## Quick start\n\n### Installation\n\n```bash\n# Default (Triton kernels)\npip install autoawq\n\n# With optimized CUDA kernels + Flash Attention\npip install autoawq[kernels]\n\n# Intel CPU\u002FXPU optimization\npip install autoawq[cpu]\n```\n\n**Requirements**: Python 3.8+, CUDA 11.8+, Compute Capability 7.5+\n\n### Load pre-quantized model\n\n```python\nfrom awq import AutoAWQForCausalLM\nfrom transformers import AutoTokenizer\n\nmodel_name = \"TheBloke\u002FMistral-7B-Instruct-v0.2-AWQ\"\n\nmodel = AutoAWQForCausalLM.from_quantized(\n    model_name,\n    fuse_layers=True  # Enable fused attention for speed\n)\ntokenizer = AutoTokenizer.from_pretrained(model_name)\n\n# Generate\ninputs = tokenizer(\"Explain quantum computing\", return_tensors=\"pt\").to(\"cuda\")\noutputs = model.generate(**inputs, max_new_tokens=200)\nprint(tokenizer.decode(outputs[0], skip_special_tokens=True))\n```\n\n### Quantize your own model\n\n```python\nfrom awq import AutoAWQForCausalLM\nfrom transformers import AutoTokenizer\n\nmodel_path = \"mistralai\u002FMistral-7B-Instruct-v0.2\"\n\n# Load model and tokenizer\nmodel = AutoAWQForCausalLM.from_pretrained(model_path)\ntokenizer = AutoTokenizer.from_pretrained(model_path)\n\n# Quantization config\nquant_config = {\n    \"zero_point\": True,      # Use zero-point quantization\n    \"q_group_size\": 128,     # Group size (128 recommended)\n    \"w_bit\": 4,              # 4-bit weights\n    \"version\": \"GEMM\"        # GEMM for batch, GEMV for single-token\n}\n\n# Quantize (uses pileval dataset by default)\nmodel.quantize(tokenizer, quant_config=quant_config)\n\n# Save\nmodel.save_quantized(\"mistral-7b-awq\")\ntokenizer.save_pretrained(\"mistral-7b-awq\")\n```\n\n**Timing**: ~10-15 min for 7B, ~1 hour for 70B models.\n\n## AWQ vs GPTQ vs bitsandbytes\n\n| Feature | AWQ | GPTQ | bitsandbytes |\n|---------|-----|------|--------------|\n| **Speedup (4-bit)** | ~2.5-3x | ~2x | ~1.5x |\n| **Accuracy loss** | \u003C5% | ~5-10% | ~5-15% |\n| **Calibration** | Minimal (128-1K tokens) | More extensive | None |\n| **Overfitting risk** | Low | Higher | N\u002FA |\n| **Best for** | Production inference | GPU inference | Easy integration |\n| **vLLM support** | Native | Yes | Limited |\n\n**Key insight**: AWQ assumes not all weights are equally important. It protects ~1% of salient weights identified by activation patterns, reducing quantization error without mixed-precision overhead.\n\n## Kernel backends\n\n### GEMM (default, batch inference)\n\n```python\nquant_config = {\n    \"zero_point\": True,\n    \"q_group_size\": 128,\n    \"w_bit\": 4,\n    \"version\": \"GEMM\"  # Best for batch sizes > 1\n}\n```\n\n### GEMV (single-token generation)\n\n```python\nquant_config = {\n    \"version\": \"GEMV\"  # 20% faster for batch_size=1\n}\n```\n\n**Limitation**: Only batch size 1, not good for large context.\n\n### Marlin (Ampere+ GPUs)\n\n```python\nfrom transformers import AwqConfig, AutoModelForCausalLM\n\nconfig = AwqConfig(\n    bits=4,\n    version=\"marlin\"  # 2x faster on A100\u002FH100\n)\n\nmodel = AutoModelForCausalLM.from_pretrained(\n    \"TheBloke\u002FMistral-7B-AWQ\",\n    quantization_config=config\n)\n```\n\n**Requirements**: Compute Capability 8.0+ (A100, H100, RTX 40xx)\n\n### ExLlamaV2 (AMD compatible)\n\n```python\nconfig = AwqConfig(\n    bits=4,\n    version=\"exllama\"  # Faster prefill, AMD GPU support\n)\n```\n\n## HuggingFace Transformers integration\n\n### Direct loading\n\n```python\nfrom transformers import AutoModelForCausalLM, AutoTokenizer\n\nmodel = AutoModelForCausalLM.from_pretrained(\n    \"TheBloke\u002Fzephyr-7B-alpha-AWQ\",\n    device_map=\"auto\"\n)\ntokenizer = AutoTokenizer.from_pretrained(\"TheBloke\u002Fzephyr-7B-alpha-AWQ\")\n```\n\n### Fused modules (recommended)\n\n```python\nfrom transformers import AwqConfig, AutoModelForCausalLM\n\nconfig = AwqConfig(\n    bits=4,\n    fuse_max_seq_len=512,  # Max sequence length for fusing\n    do_fuse=True           # Enable fused attention\u002FMLP\n)\n\nmodel = AutoModelForCausalLM.from_pretrained(\n    \"TheBloke\u002FMistral-7B-OpenOrca-AWQ\",\n    quantization_config=config\n)\n```\n\n**Note**: Fused modules cannot combine with FlashAttention2.\n\n## vLLM integration\n\n```python\nfrom vllm import LLM, SamplingParams\n\n# vLLM auto-detects AWQ models\nllm = LLM(\n    model=\"TheBloke\u002FLlama-2-7B-AWQ\",\n    quantization=\"awq\",\n    dtype=\"half\"\n)\n\nsampling = SamplingParams(temperature=0.7, max_tokens=200)\noutputs = llm.generate([\"Explain AI\"], sampling)\n```\n\n## Performance benchmarks\n\n### Memory reduction\n\n| Model | FP16 | AWQ 4-bit | Reduction |\n|-------|------|-----------|-----------|\n| Mistral 7B | 14 GB | 5.5 GB | 2.5x |\n| Llama 2-13B | 26 GB | 10 GB | 2.6x |\n| Llama 2-70B | 140 GB | 35 GB | 4x |\n\n### Inference speed (RTX 4090)\n\n| Model | Prefill (tok\u002Fs) | Decode (tok\u002Fs) | Memory |\n|-------|-----------------|----------------|--------|\n| Mistral 7B GEMM | 3,897 | 114 | 5.55 GB |\n| TinyLlama 1B GEMV | 5,179 | 431 | 2.10 GB |\n| Llama 2-13B GEMM | 2,279 | 74 | 10.28 GB |\n\n### Accuracy (perplexity)\n\n| Model | FP16 | AWQ 4-bit | Degradation |\n|-------|------|-----------|-------------|\n| Llama 3 8B | 8.20 | 8.48 | +3.4% |\n| Mistral 7B | 5.25 | 5.42 | +3.2% |\n| Qwen2 72B | 4.85 | 4.95 | +2.1% |\n\n## Custom calibration data\n\n```python\n# Use custom dataset for domain-specific models\nmodel.quantize(\n    tokenizer,\n    quant_config=quant_config,\n    calib_data=\"wikitext\",       # Or custom list of strings\n    max_calib_samples=256,       # More samples = better accuracy\n    max_calib_seq_len=512        # Sequence length\n)\n\n# Or provide your own samples\ncalib_samples = [\n    \"Your domain-specific text here...\",\n    \"More examples from your use case...\",\n]\nmodel.quantize(tokenizer, quant_config=quant_config, calib_data=calib_samples)\n```\n\n## Multi-GPU deployment\n\n```python\nmodel = AutoAWQForCausalLM.from_quantized(\n    \"TheBloke\u002FLlama-2-70B-AWQ\",\n    device_map=\"auto\",  # Auto-split across GPUs\n    max_memory={0: \"40GB\", 1: \"40GB\"}\n)\n```\n\n## Supported models\n\n35+ architectures including:\n- **Llama family**: Llama 2\u002F3, Code Llama, Mistral, Mixtral\n- **Qwen**: Qwen, Qwen2, Qwen2.5-VL\n- **Others**: Falcon, MPT, Phi, Yi, DeepSeek, Gemma\n- **Multimodal**: LLaVA, LLaVA-Next, Qwen2-VL\n\n## Common issues\n\n**CUDA OOM during quantization**:\n```python\n# Reduce batch size\nmodel.quantize(tokenizer, quant_config=quant_config, max_calib_samples=64)\n```\n\n**Slow inference**:\n```python\n# Enable fused layers\nmodel = AutoAWQForCausalLM.from_quantized(model_name, fuse_layers=True)\n```\n\n**AMD GPU support**:\n```python\n# Use ExLlama backend\nconfig = AwqConfig(bits=4, version=\"exllama\")\n```\n\n## Deprecation notice\n\nAutoAWQ is officially deprecated. For new projects, consider:\n- **vLLM llm-compressor**: https:\u002F\u002Fgithub.com\u002Fvllm-project\u002Fllm-compressor\n- **MLX-LM**: For Mac devices with Apple Silicon\n\nExisting quantized models remain usable.\n\n## References\n\n- **Paper**: AWQ: Activation-aware Weight Quantization (arXiv:2306.00978) - MLSys 2024 Best Paper\n- **GitHub**: https:\u002F\u002Fgithub.com\u002Fcasper-hansen\u002FAutoAWQ\n- **MIT Han Lab**: https:\u002F\u002Fgithub.com\u002Fmit-han-lab\u002Fllm-awq\n- **Models**: https:\u002F\u002Fhuggingface.co\u002Fmodels?library=awq\n","# AWQ (Activation-aware Weight Quantization)\n\n4-bit quantization that preserves salient weights based on activation patterns, achieving 3x speedup with minimal accuracy loss.\n\n## When to use AWQ\n\n**Use AWQ when:**\n- Need 4-bit quantization with \u003C5% accuracy loss\n- Deploying instruction-tuned or chat models (AWQ generalizes better)\n- Want ~2.5-3x inference speedup over FP16\n- Using vLLM for production serving\n- Have Ampere+ GPUs (A100, H100, RTX 40xx) for Marlin kernel support\n\n**Use GPTQ instead when:**\n- Need maximum ecosystem compatibility (more tools support GPTQ)\n- Working with ExLlamaV2 backend specifically\n- Have older GPUs without Marlin support\n\n**Use bitsandbytes instead when:**\n- Need zero calibration overhead (quantize on-the-fly)\n- Want to fine-tune with QLoRA\n- Prefer simpler integration\n\n## Quick start\n\n### Installation\n\n```bash\n# Default (Triton kernels)\npip install autoawq\n\n# With optimized CUDA kernels + Flash Attention\npip install autoawq[kernels]\n\n# Intel CPU\u002FXPU optimization\npip install autoawq[cpu]\n```\n\n**Requirements**: Python 3.8+, CUDA 11.8+, Compute Capability 7.5+\n\n### Load pre-quantized model\n\n```python\nfrom awq import AutoAWQForCausalLM\nfrom transformers import AutoTokenizer\n\nmodel_name = \"TheBloke\u002FMistral-7B-Instruct-v0.2-AWQ\"\n\nmodel = AutoAWQForCausalLM.from_quantized(\n    model_name,\n    fuse_layers=True  # Enable fused attention for speed\n)\ntokenizer = AutoTokenizer.from_pretrained(model_name)\n\n# Generate\ninputs = tokenizer(\"Explain quantum computing\", return_tensors=\"pt\").to(\"cuda\")\noutputs = model.generate(**inputs, max_new_tokens=200)\nprint(tokenizer.decode(outputs[0], skip_special_tokens=True))\n```\n\n### Quantize your own model\n\n```python\nfrom awq import AutoAWQForCausalLM\nfrom transformers import AutoTokenizer\n\nmodel_path = \"mistralai\u002FMistral-7B-Instruct-v0.2\"\n\n# Load model and tokenizer\nmodel = AutoAWQForCausalLM.from_pretrained(model_path)\ntokenizer = AutoTokenizer.from_pretrained(model_path)\n\n# Quantization config\nquant_config = {\n    \"zero_point\": True,      # Use zero-point quantization\n    \"q_group_size\": 128,     # Group size (128 recommended)\n    \"w_bit\": 4,              # 4-bit weights\n    \"version\": \"GEMM\"        # GEMM for batch, GEMV for single-token\n}\n\n# Quantize (uses pileval dataset by default)\nmodel.quantize(tokenizer, quant_config=quant_config)\n\n# Save\nmodel.save_quantized(\"mistral-7b-awq\")\ntokenizer.save_pretrained(\"mistral-7b-awq\")\n```\n\n**Timing**: ~10-15 min for 7B, ~1 hour for 70B models.\n\n## AWQ vs GPTQ vs bitsandbytes\n\n| Feature | AWQ | GPTQ | bitsandbytes |\n|---------|-----|------|--------------|\n| **Speedup (4-bit)** | ~2.5-3x | ~2x | ~1.5x |\n| **Accuracy loss** | \u003C5% | ~5-10% | ~5-15% |\n| **Calibration** | Minimal (128-1K tokens) | More extensive | None |\n| **Overfitting risk** | Low | Higher | N\u002FA |\n| **Best for** | Production inference | GPU inference | Easy integration |\n| **vLLM support** | Native | Yes | Limited |\n\n**Key insight**: AWQ assumes not all weights are equally important. It protects ~1% of salient weights identified by activation patterns, reducing quantization error without mixed-precision overhead.\n\n## Kernel backends\n\n### GEMM (default, batch inference)\n\n```python\nquant_config = {\n    \"zero_point\": True,\n    \"q_group_size\": 128,\n    \"w_bit\": 4,\n    \"version\": \"GEMM\"  # Best for batch sizes > 1\n}\n```\n\n### GEMV (single-token generation)\n\n```python\nquant_config = {\n    \"version\": \"GEMV\"  # 20% faster for batch_size=1\n}\n```\n\n**Limitation**: Only batch size 1, not good for large context.\n\n### Marlin (Ampere+ GPUs)\n\n```python\nfrom transformers import AwqConfig, AutoModelForCausalLM\n\nconfig = AwqConfig(\n    bits=4,\n    version=\"marlin\"  # 2x faster on A100\u002FH100\n)\n\nmodel = AutoModelForCausalLM.from_pretrained(\n    \"TheBloke\u002FMistral-7B-AWQ\",\n    quantization_config=config\n)\n```\n\n**Requirements**: Compute Capability 8.0+ (A100, H100, RTX 40xx)\n\n### ExLlamaV2 (AMD compatible)\n\n```python\nconfig = AwqConfig(\n    bits=4,\n    version=\"exllama\"  # Faster prefill, AMD GPU support\n)\n```\n\n## HuggingFace Transformers integration\n\n### Direct loading\n\n```python\nfrom transformers import AutoModelForCausalLM, AutoTokenizer\n\nmodel = AutoModelForCausalLM.from_pretrained(\n    \"TheBloke\u002Fzephyr-7B-alpha-AWQ\",\n    device_map=\"auto\"\n)\ntokenizer = AutoTokenizer.from_pretrained(\"TheBloke\u002Fzephyr-7B-alpha-AWQ\")\n```\n\n### Fused modules (recommended)\n\n```python\nfrom transformers import AwqConfig, AutoModelForCausalLM\n\nconfig = AwqConfig(\n    bits=4,\n    fuse_max_seq_len=512,  # Max sequence length for fusing\n    do_fuse=True           # Enable fused attention\u002FMLP\n)\n\nmodel = AutoModelForCausalLM.from_pretrained(\n    \"TheBloke\u002FMistral-7B-OpenOrca-AWQ\",\n    quantization_config=config\n)\n```\n\n**Note**: Fused modules cannot combine with FlashAttention2.\n\n## vLLM integration\n\n```python\nfrom vllm import LLM, SamplingParams\n\n# vLLM auto-detects AWQ models\nllm = LLM(\n    model=\"TheBloke\u002FLlama-2-7B-AWQ\",\n    quantization=\"awq\",\n    dtype=\"half\"\n)\n\nsampling = SamplingParams(temperature=0.7, max_tokens=200)\noutputs = llm.generate([\"Explain AI\"], sampling)\n```\n\n## Performance benchmarks\n\n### Memory reduction\n\n| Model | FP16 | AWQ 4-bit | Reduction |\n|-------|------|-----------|-----------|\n| Mistral 7B | 14 GB | 5.5 GB | 2.5x |\n| Llama 2-13B | 26 GB | 10 GB | 2.6x |\n| Llama 2-70B | 140 GB | 35 GB | 4x |\n\n### Inference speed (RTX 4090)\n\n| Model | Prefill (tok\u002Fs) | Decode (tok\u002Fs) | Memory |\n|-------|-----------------|----------------|--------|\n| Mistral 7B GEMM | 3,897 | 114 | 5.55 GB |\n| TinyLlama 1B GEMV | 5,179 | 431 | 2.10 GB |\n| Llama 2-13B GEMM | 2,279 | 74 | 10.28 GB |\n\n### Accuracy (perplexity)\n\n| Model | FP16 | AWQ 4-bit | Degradation |\n|-------|------|-----------|-------------|\n| Llama 3 8B | 8.20 | 8.48 | +3.4% |\n| Mistral 7B | 5.25 | 5.42 | +3.2% |\n| Qwen2 72B | 4.85 | 4.95 | +2.1% |\n\n## Custom calibration data\n\n```python\n# Use custom dataset for domain-specific models\nmodel.quantize(\n    tokenizer,\n    quant_config=quant_config,\n    calib_data=\"wikitext\",       # Or custom list of strings\n    max_calib_samples=256,       # More samples = better accuracy\n    max_calib_seq_len=512        # Sequence length\n)\n\n# Or provide your own samples\ncalib_samples = [\n    \"Your domain-specific text here...\",\n    \"More examples from your use case...\",\n]\nmodel.quantize(tokenizer, quant_config=quant_config, calib_data=calib_samples)\n```\n\n## Multi-GPU deployment\n\n```python\nmodel = AutoAWQForCausalLM.from_quantized(\n    \"TheBloke\u002FLlama-2-70B-AWQ\",\n    device_map=\"auto\",  # Auto-split across GPUs\n    max_memory={0: \"40GB\", 1: \"40GB\"}\n)\n```\n\n## Supported models\n\n35+ architectures including:\n- **Llama family**: Llama 2\u002F3, Code Llama, Mistral, Mixtral\n- **Qwen**: Qwen, Qwen2, Qwen2.5-VL\n- **Others**: Falcon, MPT, Phi, Yi, DeepSeek, Gemma\n- **Multimodal**: LLaVA, LLaVA-Next, Qwen2-VL\n\n## Common issues\n\n**CUDA OOM during quantization**:\n```python\n# Reduce batch size\nmodel.quantize(tokenizer, quant_config=quant_config, max_calib_samples=64)\n```\n\n**Slow inference**:\n```python\n# Enable fused layers\nmodel = AutoAWQForCausalLM.from_quantized(model_name, fuse_layers=True)\n```\n\n**AMD GPU support**:\n```python\n# Use ExLlama backend\nconfig = AwqConfig(bits=4, version=\"exllama\")\n```\n\n## Deprecation notice\n\nAutoAWQ is officially deprecated. For new projects, consider:\n- **vLLM llm-compressor**: https:\u002F\u002Fgithub.com\u002Fvllm-project\u002Fllm-compressor\n- **MLX-LM**: For Mac devices with Apple Silicon\n\nExisting quantized models remain usable.\n\n## References\n\n- **Paper**: AWQ: Activation-aware Weight Quantization (arXiv:2306.00978) - MLSys 2024 Best Paper\n- **GitHub**: https:\u002F\u002Fgithub.com\u002Fcasper-hansen\u002FAutoAWQ\n- **MIT Han Lab**: https:\u002F\u002Fgithub.com\u002Fmit-han-lab\u002Fllm-awq\n- **Models**: https:\u002F\u002Fhuggingface.co\u002Fmodels?library=awq\n",{"name":5,"tags":49,"author":50,"license":15,"version":51,"description":7,"dependencies":52},[42,26,25,37,27,35,41,34,33],"Orchestra Research","1.0.0",[53,54,55],"autoawq","transformers>=4.45.0","torch>=2.0.0","main","75c7c92896e878938460f0bceb23b6ab4a787286","AI-Research-SKILLs\u002F10-optimization\u002Fawq","https:\u002F\u002Fgithub.com\u002Fzhengliangliahu-cloud\u002Fskills","https:\u002F\u002Fgithub.com\u002Fzhengliangliahu-cloud","A postgraduate student majoring in mathematics.","2026-06-03 19:04:43",{"source":8,"skillName":5,"commands":64,"prompt":74,"zipAvailable":75},[65,68,71],{"tool":66,"command":67},"npx","npx skills add zhengliangliahu-cloud\u002Fskills --skill awq-quantization",{"tool":69,"command":70},"bunx","bunx skills add zhengliangliahu-cloud\u002Fskills --skill awq-quantization",{"tool":72,"command":73},"pnpm","pnpm dlx skills add zhengliangliahu-cloud\u002Fskills --skill awq-quantization","安装技能，技能地址：https:\u002F\u002Fgithub.com\u002Fzhengliangliahu-cloud\u002Fskills\u002Fblob\u002Fmain\u002FAI-Research-SKILLs\u002F10-optimization\u002Fawq\u002FSKILL.md",true,[77,83],{"path":78,"resourceType":79,"resourceTypeDesc":80,"sizeBytes":81,"sourceHtmlUrl":82},"references\u002Fadvanced-usage.md","REFERENCE","参考文档",7983,"https:\u002F\u002Fgithub.com\u002Fzhengliangliahu-cloud\u002Fskills\u002Fblob\u002Fmain\u002FAI-Research-SKILLs\u002F10-optimization\u002Fawq\u002Freferences\u002Fadvanced-usage.md",{"path":84,"resourceType":79,"resourceTypeDesc":80,"sizeBytes":85,"sourceHtmlUrl":86},"references\u002Ftroubleshooting.md",7733,"https:\u002F\u002Fgithub.com\u002Fzhengliangliahu-cloud\u002Fskills\u002Fblob\u002Fmain\u002FAI-Research-SKILLs\u002F10-optimization\u002Fawq\u002Freferences\u002Ftroubleshooting.md",[88,116,150,174,199,219],{"skillNo":89,"name":90,"nameZh":6,"description":91,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":92,"categories":93,"keywords":95,"publishedAt":44},"01M6NBV8V36BCHZ1K9F7EMRJP1","ara-rigor-reviewer","Performs ARA Seal Level 2 semantic epistemic review on Agent-Native Research Artifacts, scoring six dimensions (evidence relevance, falsifiability, scope calibration, argument coherence, exploration integrity, methodological rigor) and producing a constructive, severity-ranked report with a Strong Accept-to-Reject recommendation. Use after Level 1 structural validation passes, when an ARA needs an objective epistemic critique before publication or release.","https:\u002F\u002Fgithub.com\u002Fzhengliangliahu-cloud\u002Fskills\u002Fblob\u002Fmain\u002FAI-Research-SKILLs\u002F22-agent-native-research-artifact\u002Frigor-reviewer\u002FSKILL.md",[94],{"categoryNo":19,"categoryName":20,"taxonomyType":21,"taxonomyTypeDesc":22},[96,97,98,99,100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115],"reviewer","ara rigor reviewer","ARA","review","release","audit","integrity","peer review","critique","exploration","scoring","dimensions","coherence","structural","scope","native","argument","epistemic","calibration","Research Artifacts",{"skillNo":117,"name":118,"nameZh":6,"description":119,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":120,"categories":121,"keywords":131,"publishedAt":44},"01M6NBTQTZ5G21FFPMEKHQZ3R3","ara-research-manager","Records research provenance as a post-task epilogue, scanning conversation history at the end of a coding or research session to extract decisions, experiments, dead ends, claims, heuristics, and pivots, and writing them into the ara\u002F directory with user-vs-AI provenance tags. Use as a session epilogue — never during execution — to maintain a faithful, auditable trace of how a research project actually evolved.","https:\u002F\u002Fgithub.com\u002Fzhengliangliahu-cloud\u002Fskills\u002Fblob\u002Fmain\u002FAI-Research-SKILLs\u002F22-agent-native-research-artifact\u002Fresearch-manager\u002FSKILL.md",[122,125,126],{"categoryNo":123,"categoryName":124,"taxonomyType":21,"taxonomyTypeDesc":22},"01JBSEEDCAT000000000000018","写作与文案",{"categoryNo":19,"categoryName":20,"taxonomyType":21,"taxonomyTypeDesc":22},{"categoryNo":127,"categoryName":128,"taxonomyType":129,"taxonomyTypeDesc":130},"01JBSEEDCAT000000000000036","内容创作者","PROFESSION","适用职业",[132,133,134,135,136,137,138,139,140,141,142,143,105,144,145,146,147,148,149,98],"research","manager","ara research manager","writing","directory","heuristics","how","knowledge","knowledge management","never","scanning","conversation","maintain","project","end","recording","at","claims",{"skillNo":151,"name":152,"nameZh":6,"description":153,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":154,"categories":155,"keywords":160,"publishedAt":44},"01M6NBTCRK24WXZYKBD19RAV09","ara-compiler","Compiles any research input — PDF papers, GitHub repositories, experiment logs, code directories, or raw notes — into a complete Agent-Native Research Artifact (ARA) with cognitive layer (claims, concepts, heuristics), physical layer (configs, code stubs), exploration graph, and grounded evidence. Use when ingesting a paper or codebase into a structured, machine-executable knowledge package, building an ARA from scratch, or converting research outputs into a falsifiable, agent-traversable form.","https:\u002F\u002Fgithub.com\u002Fzhengliangliahu-cloud\u002Fskills\u002Fblob\u002Fmain\u002FAI-Research-SKILLs\u002F22-agent-native-research-artifact\u002Fcompiler\u002FSKILL.md",[156,159],{"categoryNo":157,"categoryName":158,"taxonomyType":21,"taxonomyTypeDesc":22},"01JBSEEDCAT000000000000002","文档处理",{"categoryNo":19,"categoryName":20,"taxonomyType":21,"taxonomyTypeDesc":22},[161,162,163,164,165,166,137,139,167,105,168,169,111,170,171,172,149,98,115,173],"compiler","ara compiler","PDF","notes","code","logs","ingestion","artifact","input","papers","directories","layer","Knowledge Extraction",{"skillNo":175,"name":176,"nameZh":6,"description":177,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":178,"categories":179,"keywords":181,"publishedAt":44},"01M6NBSXGH49PE13PF1H9P7F8N","creative-thinking-for-research","Applies cognitive science frameworks for creative thinking to CS and AI research ideation. Use when seeking genuinely novel research directions by leveraging combinatorial creativity, analogical reasoning, constraint manipulation, and other empirically grounded creative strategies.","https:\u002F\u002Fgithub.com\u002Fzhengliangliahu-cloud\u002Fskills\u002Fblob\u002Fmain\u002FAI-Research-SKILLs\u002F21-research-ideation\u002Fcreative-thinking-for-research\u002FSKILL.md",[180],{"categoryNo":19,"categoryName":20,"taxonomyType":21,"taxonomyTypeDesc":22},[132,182,183,184,185,29,186,187,188,189,190,191,192,193,194,195,196,197,30,198],"creative","thinking","Creative Thinking","creative thinking for research","reasoning","science","use","ideation","frameworks","research ideation","strategies","problem","Analogical Reasoning","Problem Reformulation","novel","directions","constraint",{"skillNo":200,"name":201,"nameZh":6,"description":202,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":203,"categories":204,"keywords":206,"publishedAt":44},"01M6NBSNHZYM70ZAFXQNESYYBG","brainstorming-research-ideas","Guides researchers through structured ideation frameworks to discover high-impact research directions. Use when exploring new problem spaces, pivoting between projects, or seeking novel angles on existing work.","https:\u002F\u002Fgithub.com\u002Fzhengliangliahu-cloud\u002Fskills\u002Fblob\u002Fmain\u002FAI-Research-SKILLs\u002F21-research-ideation\u002Fbrainstorming-research-ideas\u002FSKILL.md",[205],{"categoryNo":19,"categoryName":20,"taxonomyType":21,"taxonomyTypeDesc":22},[132,207,208,209,210,211,212,213,183,190,214,193,215,216,184,217,218,196,197,30],"brainstorming","ideas","brainstorming research ideas","discovery","projects","work","strategy","discover","impact","guides","Problem Discovery","Research Strategy",{"skillNo":220,"name":221,"nameZh":6,"description":222,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":223,"categories":224,"keywords":231,"publishedAt":44},"01M6NBS3Z3CJYH5225X73AZN3S","systems-paper-writing","Comprehensive guide for writing systems papers targeting OSDI, SOSP, ASPLOS, NSDI, and EuroSys. Provides paragraph-level structural blueprints, writing patterns, venue-specific checklists, reviewer guidelines, LaTeX templates, and conference deadlines. Use this skill for all systems conference paper writing.","https:\u002F\u002Fgithub.com\u002Fzhengliangliahu-cloud\u002Fskills\u002Fblob\u002Fmain\u002FAI-Research-SKILLs\u002F20-ml-paper-writing\u002Fsystems-paper-writing\u002FSKILL.md",[225,226,227,230],{"categoryNo":123,"categoryName":124,"taxonomyType":21,"taxonomyTypeDesc":22},{"categoryNo":19,"categoryName":20,"taxonomyType":21,"taxonomyTypeDesc":22},{"categoryNo":228,"categoryName":229,"taxonomyType":21,"taxonomyTypeDesc":22},"01JBSEEDCAT000000000000023","Skill与智能体开发",{"categoryNo":127,"categoryName":128,"taxonomyType":129,"taxonomyTypeDesc":130},[135,232,233,234,235,96,236,237,238,109,239,240,241,242,170,243,30,29,244,245],"Paper Writing","systems","paper","Systems Paper Writing","guide","Blueprint","patterns","OSDI","NSDI","ASPLOS","SOSP","targeting","EuroSys","Structural Blueprint"]