{"_id":"@avasis-ai/inspect","_rev":"2-a0667e66d68b47dbf0601a67a0bc3999","name":"@avasis-ai/inspect","dist-tags":{"latest":"1.0.1"},"versions":{"1.0.0":{"name":"@avasis-ai/inspect","version":"1.0.0","keywords":["avasis","agent","inspect","hardware","ollama","llm","local-ai"],"license":"MIT","_id":"@avasis-ai/inspect@1.0.0","maintainers":[{"name":"avasis","email":"abhay@avasis.ai"}],"bin":{"avasis-inspect":"dist/bin.js"},"dist":{"shasum":"c2b0ce48e2611d98665a5d956cd4bb1fb40d14de","tarball":"https://registry.npmjs.org/@avasis-ai/inspect/-/inspect-1.0.0.tgz","fileCount":10,"integrity":"sha512-sBLSsJUkd1FVOn4p1BeaxPPHnMpUiB4U/bKIzweCDdEKh9fcZKLRKI1yVAg6uzbg5yyOAgVApWK2Mw3+G9lzSA==","signatures":[{"sig":"MEUCIEzyzra/a5MlpuyxaJ9ua5N36aG4wTtnpmZghkZ6z2Q8AiEA07PLVKgrchzu7WLiT+HV+6lKQjsCpUeHOvH6MWN9oB0=","keyid":"SHA256:DhQ8wR5APBvFHLF/+Tc+AYvPOdTpcIDqOhxsBHRwC7U"}],"unpackedSize":18972},"type":"module","engines":{"node":">=18"},"scripts":{"build":"tsc","prepublishOnly":"npm run build"},"_npmUser":{"name":"avasis","email":"abhay@avasis.ai"},"_npmVersion":"11.7.0","description":"Hardware inspector for Avasis Agent Builder. Detects CPU, RAM, GPU and recommends small models that run on your device.","directories":{},"_nodeVersion":"25.4.0","dependencies":{"systeminformation":"^5.23.0"},"_hasShrinkwrap":false,"devDependencies":{"typescript":"^5.7.0","@types/node":"^22.0.0"},"_npmOperationalInternal":{"tmp":"tmp/inspect_1.0.0_1775679012951_0.320419057373428","host":"s3://npm-registry-packages-npm-production"}},"1.0.1":{"name":"@avasis-ai/inspect","version":"1.0.1","description":"Hardware inspector for Avasis Agent Builder. Detects CPU, RAM, GPU and recommends small models that run on your device.","type":"module","bin":{"avasis-inspect":"dist/bin.js"},"scripts":{"build":"tsc","prepublishOnly":"npm run build"},"keywords":["avasis","agent","inspect","hardware","ollama","llm","local-ai"],"license":"MIT","engines":{"node":">=18"},"dependencies":{"systeminformation":"^5.23.0"},"devDependencies":{"@types/node":"^22.0.0","typescript":"^5.7.0"},"_id":"@avasis-ai/inspect@1.0.1","_nodeVersion":"25.4.0","_npmVersion":"11.7.0","dist":{"integrity":"sha512-EZ7bly9Uvp4a9+FMpYYKmDl9c03Coqdyipi4rcDubRlHcMju29oy62Cs9eBn3FgU4W2Zc2rwuS2eF07/8fZtmQ==","shasum":"7f432c9197ccb914dec885ebc0d98406350074a7","tarball":"https://registry.npmjs.org/@avasis-ai/inspect/-/inspect-1.0.1.tgz","fileCount":10,"unpackedSize":19002,"signatures":[{"keyid":"SHA256:DhQ8wR5APBvFHLF/+Tc+AYvPOdTpcIDqOhxsBHRwC7U","sig":"MEYCIQC31mboF/UHEj81KoNyA2isFa7mKpeZ7L4ST76B0YSpYQIhAMsqEwmajbK07Hbkuo1N3Y0kLeBjtSOeZ6KTMEATTcXj"}]},"_npmUser":{"name":"avasis","email":"abhay@avasis.ai"},"directories":{},"maintainers":[{"name":"avasis","email":"abhay@avasis.ai"}],"_npmOperationalInternal":{"host":"s3://npm-registry-packages-npm-production","tmp":"tmp/inspect_1.0.1_1775679079917_0.9074746530912803"},"_hasShrinkwrap":false}},"time":{"created":"2026-04-08T20:10:12.885Z","modified":"2026-04-08T20:11:20.184Z","1.0.0":"2026-04-08T20:10:13.117Z","1.0.1":"2026-04-08T20:11:20.075Z"},"license":"MIT","keywords":["avasis","agent","inspect","hardware","ollama","llm","local-ai"],"description":"Hardware inspector for Avasis Agent Builder. Detects CPU, RAM, GPU and recommends small models that run on your device.","maintainers":[{"name":"avasis","email":"abhay@avasis.ai"}],"readme":"# @avasis-ai/inspect\n\nHardware inspector for Avasis Agent Builder. Scans your device and recommends the best small models that run locally.\n\n## Install\n\n```bash\nnpx @avasis-ai/inspect\n```\n\n## What it does\n\n- Detects CPU, RAM, GPU (VRAM), OS, and disk\n- Matches your hardware against a catalog of 12 small models (0.6B to 14B)\n- Recommends models that fit your device with GPU-accelerated or CPU-only ratings\n- Outputs a Config ID you can use to pre-fill the agent builder at avasis.ai\n\n## Usage\n\n```bash\n# Interactive scan with colored output\nnpx @avasis-ai/inspect\n\n# JSON output (for automation or builder pre-fill)\nnpx @avasis-ai/inspect --json\n```\n\n## Example output\n\n```\n  ┌─────────────────────────────────────────────┐\n  │        Avasis Device Inspector              │\n  └─────────────────────────────────────────────┘\n\n  DEVICE\n  ─────────────────────────────────────────────\n  CPU     Apple M1 Pro\n  Cores   10 physical / 10 logical\n  RAM     32.0GB total, 18.2GB free\n  OS      macOS 15.0 (arm64)\n  Disk    456.2GB free of 960.0GB\n\n  RECOMMENDED MODELS\n  ─────────────────────────────────────────────\n  Model                 Size      Speed     Quality   Fit\n  ──────────────────────────────────────────────────────────────\n  qwen3:0.6b            0.4GB Q4  fast      basic     CPU\n  phi4-mini:3.8b        2.4GB Q4  fast      good      CPU\n  gemma3:4b             2.8GB Q4  fast      good      CPU\n  ...\n\n  BEST MODEL FOR YOUR DEVICE\n  ─────────────────────────────────────────────\n  phi4-mini:3.8b\n  No GPU detected, runs on CPU\n\n  NEXT STEPS\n  ─────────────────────────────────────────────\n  1. Install Ollama:     https://ollama.com/download\n  2. Pull your model:    ollama pull phi4-mini:3.8b\n  3. Build an agent:     https://avasis.ai/builder\n\n  Config ID: a1b2c3d4 (use this to pre-fill your agent)\n```\n\n## Supported models\n\n| Model | Parameters | Quantized Size | Min RAM | Best For |\n|-------|-----------|----------------|---------|----------|\n| qwen3:0.6b | 0.6B | 0.4GB Q4 | 2GB | Classification, simple Q&A |\n| phi4-mini:3.8b | 3.8B | 2.4GB Q4 | 4GB | Code assist, file tasks |\n| gemma3:4b | 4B | 2.8GB Q4 | 6GB | Code generation, multi-tool agents |\n| qwen3:4b | 4B | 2.6GB Q4 | 6GB | Reasoning, structured output |\n| llama3.2:3b | 3.2B | 2.0GB Q4 | 4GB | Instruction following, tool calling |\n| mistral:7b | 7B | 4.4GB Q4 | 8GB | Complex reasoning, code generation |\n| qwen3:8b | 8B | 5.0GB Q4 | 8GB | Deep analysis, multi-agent supervisor |\n| phi4:14b | 14B | 8.5GB Q4 | 16GB | Expert reasoning, full autonomy |\n| qwen3:14b | 14B | 9.0GB Q4 | 16GB | Research, long-context tasks |\n| gemma3:12b | 12B | 7.5GB Q4 | 16GB | Multi-modal, supervisor agent |\n| llama3.1:8b | 8B | 4.9GB Q4 | 8GB | General purpose, agentic tasks |\n| deepseek-r1:7b | 7B | 4.4GB Q4 | 8GB | Reasoning, math, logic |\n\n## How it works\n\n1. Uses `systeminformation` to scan hardware (cross-platform, no native deps)\n2. Compares your specs against a model catalog with real size/requirement data\n3. Rates each model as GPU-native, CPU-only, or incompatible\n4. Picks the best model for your device\n\n## License\n\nMIT\n","readmeFilename":"README.md"}