Server definition

Hash: sha256:43211133cf932dd14ee2bc095ae03e5ed43c1dba9e6af8d9f64ae6a629d26503
What it is: What a remote MCP server returned when asked what it offers: 9 tools

The blob, as servednamed by its sha256

{
  "instructions": null,
  "tools": [
    {
      "description": "Will a given local LLM run on given hardware? Returns fit, the best quant that fits, theoretical tok/s, and real owner-measured tok/s where available.",
      "inputSchema": {
        "$schema": "http://json-schema.org/draft-07/schema#",
        "additionalProperties": false,
        "properties": {
          "active_b": {
            "description": "For an unlisted model: active params in billions (= total for dense, less for MoE)",
            "type": "number"
          },
          "bandwidth_gbps": {
            "description": "For custom hardware: memory bandwidth in GB/s",
            "type": "number"
          },
          "context": {
            "description": "Context window in tokens (default 8192)",
            "type": "number"
          },
          "hardware": {
            "description": "Hardware name/id, e.g. 'rtx-3090', 'Mac 128GB', 'Strix Halo'. Use list_hardware to see known ones.",
            "type": "string"
          },
          "kv_precision": {
            "description": "KV cache precision (default f16)",
            "enum": [
              "f16",
              "q8",
              "q4"
            ],
            "type": "string"
          },
          "model": {
            "description": "Model name, e.g. 'Llama 70B', 'gpt-oss-120B', 'Qwen 32B'. Use list_models to see known names.",
            "type": "string"
          },
          "mxfp4": {
            "description": "True if the model ships natively in MXFP4 (e.g. gpt-oss)",
            "type": "boolean"
          },
          "total_b": {
            "description": "For an unlisted model: total parameters in billions",
            "type": "number"
          },
          "unified": {
            "description": "True for unified-memory machines (Macs, Strix Halo, CPU+RAM)",
            "type": "boolean"
          },
          "vram_gb": {
            "description": "For custom hardware: VRAM or unified memory in GB",
            "type": "number"
          }
        },
        "type": "object"
      },
      "name": "can_i_run_it",
      "outputSchema": null
    },
    {
      "description": "The cheapest catalogued, buyable machine that runs a given model at Q4 with the requested context.",
      "inputSchema": {
        "$schema": "http://json-schema.org/draft-07/schema#",
        "additionalProperties": false,
        "properties": {
          "active_b": {
            "description": "For an unlisted model: active params in billions (= total for dense, less for MoE)",
            "type": "number"
          },
          "context": {
            "description": "Context window in tokens (default 8192)",
            "type": "number"
          },
          "model": {
            "description": "Model name, e.g. 'Llama 70B', 'gpt-oss-120B', 'Qwen 32B'. Use list_models to see known names.",
            "type": "string"
          },
          "mxfp4": {
            "description": "True if the model ships natively in MXFP4 (e.g. gpt-oss)",
            "type": "boolean"
          },
          "total_b": {
            "description": "For an unlisted model: total parameters in billions",
            "type": "number"
          }
        },
        "type": "object"
      },
      "name": "cheapest_hardware_for_model",
      "outputSchema": null
    },
    {
      "description": "Side-by-side memory, bandwidth, price, and (with a model) fit + tok/s for 2 to 4 machines.",
      "inputSchema": {
        "$schema": "http://json-schema.org/draft-07/schema#",
        "additionalProperties": false,
        "properties": {
          "active_b": {
            "description": "For an unlisted model: active params in billions (= total for dense, less for MoE)",
            "type": "number"
          },
          "context": {
            "description": "Context window in tokens (default 8192)",
            "type": "number"
          },
          "hardware": {
            "description": "2 to 4 hardware names/ids, comma-separated",
            "type": "string"
          },
          "kv_precision": {
            "description": "KV cache precision (default f16)",
            "enum": [
              "f16",
              "q8",
              "q4"
            ],
            "type": "string"
          },
          "model": {
            "description": "Model name, e.g. 'Llama 70B', 'gpt-oss-120B', 'Qwen 32B'. Use list_models to see known names.",
            "type": "string"
          },
          "mxfp4": {
            "description": "True if the model ships natively in MXFP4 (e.g. gpt-oss)",
            "type": "boolean"
          },
          "total_b": {
            "description": "For an unlisted model: total parameters in billions",
            "type": "number"
          }
        },
        "required": [
          "hardware"
        ],
        "type": "object"
      },
      "name": "compare_hardware",
      "outputSchema": null
    },
    {
      "description": "Buy vs rent vs API cost to run a model locally: monthly/1y/3y totals, break-even months, and the energy cost per 1M tokens. Same math as /cost-calculator/.",
      "inputSchema": {
        "$schema": "http://json-schema.org/draft-07/schema#",
        "additionalProperties": false,
        "properties": {
          "api": {
            "description": "API $/million tokens (default 1.0)",
            "type": "number"
          },
          "hardware": {
            "description": "Catalogued hardware name/id (see list_hardware), e.g. 'rtx-3090-used'",
            "type": "string"
          },
          "hours": {
            "description": "Active hours per day (default 3)",
            "type": "number"
          },
          "kwh": {
            "description": "Electricity $/kWh (default 0.16)",
            "type": "number"
          },
          "price_usd": {
            "description": "For custom hardware: price in USD",
            "type": "number"
          },
          "rent": {
            "description": "Cloud GPU $/hour (default 0.59)",
            "type": "number"
          },
          "tdp_w": {
            "description": "For custom hardware: board power draw in watts",
            "type": "number"
          },
          "tokens": {
            "description": "Tokens generated per day, for the API comparison (default 300000)",
            "type": "number"
          }
        },
        "type": "object"
      },
      "name": "cost_compare",
      "outputSchema": null
    },
    {
      "description": "Current typical used-GPU prices for local-AI rigs (eBay Browse API median asking + hand-verified, monthly).",
      "inputSchema": {
        "$schema": "http://json-schema.org/draft-07/schema#",
        "additionalProperties": false,
        "properties": {
          "gpu": {
            "description": "Optional name/id filter, e.g. \"3090\"",
            "type": "string"
          }
        },
        "type": "object"
      },
      "name": "get_used_gpu_prices",
      "outputSchema": null
    },
    {
      "description": "List the machines the tools know about (memory, bandwidth, price, buy link).",
      "inputSchema": {
        "$schema": "http://json-schema.org/draft-07/schema#",
        "properties": {},
        "type": "object"
      },
      "name": "list_hardware",
      "outputSchema": null
    },
    {
      "description": "List the local LLM model classes the tools know about (params, dense/MoE, native context).",
      "inputSchema": {
        "$schema": "http://json-schema.org/draft-07/schema#",
        "properties": {},
        "type": "object"
      },
      "name": "list_models",
      "outputSchema": null
    },
    {
      "description": "Ranked list of catalogued, buyable machines that run a model at the requested context, cheapest first, with an optional budget cap.",
      "inputSchema": {
        "$schema": "http://json-schema.org/draft-07/schema#",
        "additionalProperties": false,
        "properties": {
          "active_b": {
            "description": "For an unlisted model: active params in billions (= total for dense, less for MoE)",
            "type": "number"
          },
          "budget": {
            "description": "Optional max price in USD",
            "type": "number"
          },
          "context": {
            "description": "Context window in tokens (default 8192)",
            "type": "number"
          },
          "kv_precision": {
            "description": "KV cache precision (default f16)",
            "enum": [
              "f16",
              "q8",
              "q4"
            ],
            "type": "string"
          },
          "model": {
            "description": "Model name, e.g. 'Llama 70B', 'gpt-oss-120B', 'Qwen 32B'. Use list_models to see known names.",
            "type": "string"
          },
          "mxfp4": {
            "description": "True if the model ships natively in MXFP4 (e.g. gpt-oss)",
            "type": "boolean"
          },
          "total_b": {
            "description": "For an unlisted model: total parameters in billions",
            "type": "number"
          }
        },
        "type": "object"
      },
      "name": "recommend_hardware",
      "outputSchema": null
    },
    {
      "description": "Which GGUF quantization to download for a model on given hardware: the full quant ladder with file size, max context, and tok/s for each, plus the recommended pick.",
      "inputSchema": {
        "$schema": "http://json-schema.org/draft-07/schema#",
        "additionalProperties": false,
        "properties": {
          "active_b": {
            "description": "For an unlisted model: active params in billions (= total for dense, less for MoE)",
            "type": "number"
          },
          "bandwidth_gbps": {
            "description": "For custom hardware: memory bandwidth in GB/s",
            "type": "number"
          },
          "context": {
            "description": "Context window in tokens (default 8192)",
            "type": "number"
          },
          "hardware": {
            "description": "Hardware name/id, e.g. 'rtx-3090', 'Mac 128GB', 'Strix Halo'. Use list_hardware to see known ones.",
            "type": "string"
          },
          "kv_precision": {
            "description": "KV cache precision (default f16)",
            "enum": [
              "f16",
              "q8",
              "q4"
            ],
            "type": "string"
          },
          "model": {
            "description": "Model name, e.g. 'Llama 70B', 'gpt-oss-120B', 'Qwen 32B'. Use list_models to see known names.",
            "type": "string"
          },
          "mxfp4": {
            "description": "True if the model ships natively in MXFP4 (e.g. gpt-oss)",
            "type": "boolean"
          },
          "total_b": {
            "description": "For an unlisted model: total parameters in billions",
            "type": "number"
          },
          "unified": {
            "description": "True for unified-memory machines (Macs, Strix Halo, CPU+RAM)",
            "type": "boolean"
          },
          "vram_gb": {
            "description": "For custom hardware: VRAM or unified memory in GB",
            "type": "number"
          }
        },
        "type": "object"
      },
      "name": "recommend_quant",
      "outputSchema": null
    }
  ]
}

Verify it yourself

curl -s https://api.teppi.xyz/v1/evidence/sha256:43211133cf932dd14ee2bc095ae03e5ed43c1dba9e6af8d9f64ae6a629d26503 | sha256sum