addrbookFor publishers
← The evidence ledger

Evidence row 574629

Observed , by probe.mcp/1.

Kind
mcp.tools
About
endpoint bbcc6c10-0230-43fd-85c8-34d0726851bf
Recorded
Previous row
None: the first of its kind about this subject
Body SHA-256
19a9a714d7bdd5681216d0579e21ef33bd10513488f4d5a4dfb80a4c20052799
Leaf
0x27d2d2b3c1d60f3065100648c972a128821e3158b2c04b83cc0896838a68d907

What was observed

{
  "count": 9,
  "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"
    },
    {
      "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"
    },
    {
      "description": "Side-by-side memory, bandwidth, price, and (with a model) fit + tok/s for 2 to 4 machines.",
      "inputSchema": {
        "$schema": "http://json-schema.org/draft-07/schema#",
        "additionalProperties": false,
        "properties": {
          "active_b": {
            "description": "For an unlisted model: active params in billions (= total for dense, less for MoE)",
            "type": "number"
          },
          "context": {
            "description": "Context window in tokens (default 8192)",
            "type": "number"
          },
          "hardware": {
            "description": "2 to 4 hardware names/ids, comma-separated",
            "type": "string"
          },
          "kv_precision": {
            "description": "KV cache precision (default f16)",
            "enum": [
              "f16",
              "q8",
              "q4"
            ],
            "type": "string"
          },
          "model": {
            "description": "Model name, e.g. 'Llama 70B', 'gpt-oss-120B', 'Qwen 32B'. Use list_models to see known names.",
            "type": "string"
          },
          "mxfp4": {
            "description": "True if the model ships natively in MXFP4 (e.g. gpt-oss)",
            "type": "boolean"
          },
          "total_b": {
            "description": "For an unlisted model: total parameters in billions",
            "type": "number"
          }
        },
        "required": [
          "hardware"
        ],
        "type": "object"
      },
      "name": "compare_hardware"
    },
    {
      "description": "Buy vs rent vs API cost to run a model locally: monthly/1y/3y totals, break-even months, and the energy cost per 1M tokens. Same math as /cost-calculator/.",
      "inputSchema": {
        "$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 $/hour (default 0.59)",
            "type": "number"
          },
          "tdp_w": {
            "description": "For custom hardware: board power draw in watts",
            "type": "number"
          },
          "tokens": {
            "description": "Tokens generated per day, for the API comparison (default 300000)",
            "type": "number"
          }
        },
        "type": "object"
      },
      "name": "cost_compare"
    },
    {
      "description": "Current typical used-GPU prices for local-AI rigs (eBay Browse API median asking + hand-verified, monthly).",
      "inputSchema": {
        "$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"
    },
    {
      "description": "List the machines the tools know about (memory, bandwidth, price, buy link).",
      "inputSchema": {
        "$schema": "http://json-schema.org/draft-07/schema#",
        "properties": {},
        "type": "object"
      },
      "name": "list_hardware"
    },
    {
      "description": "List the local LLM model classes the tools know about (params, dense/MoE, native context).",
      "inputSchema": {
        "$schema": "http://json-schema.org/draft-07/schema#",
        "properties": {},
        "type": "object"
      },
      "name": "list_models"
    },
    {
      "description": "Ranked list of catalogued, buyable machines that run a model at the requested context, cheapest first, with an optional budget cap.",
      "inputSchema": {
        "$schema": "http://json-schema.org/draft-07/schema#",
        "additionalProperties": false,
        "properties": {
          "active_b": {
            "description": "For an unlisted model: active params in billions (= total for dense, less for MoE)",
            "type": "number"
          },
          "budget": {
            "description": "Optional max price in USD",
            "type": "number"
          },
          "context": {
            "description": "Context window in tokens (default 8192)",
            "type": "number"
          },
          "kv_precision": {
            "description": "KV cache precision (default f16)",
            "enum": [
              "f16",
              "q8",
              "q4"
            ],
            "type": "string"
          },
          "model": {
            "description": "Model name, e.g. 'Llama 70B', 'gpt-oss-120B', 'Qwen 32B'. Use list_models to see known names.",
            "type": "string"
          },
          "mxfp4": {
            "description": "True if the model ships natively in MXFP4 (e.g. gpt-oss)",
            "type": "boolean"
          },
          "total_b": {
            "description": "For an unlisted model: total parameters in billions",
            "type": "number"
          }
        },
        "type": "object"
      },
      "name": "recommend_hardware"
    },
    {
      "description": "Which GGUF quantization to download for a model on given hardware: the full quant ladder with file size, max context, and tok/s for each, plus the recommended pick.",
      "inputSchema": {
        "$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"
    }
  ]
}

What the leaf hashes

{"body":"19a9a714d7bdd5681216d0579e21ef33bd10513488f4d5a4dfb80a4c20052799","id":574629,"kind":"mcp.tools","method":"probe.mcp/1","observed_us":1793141597762751,"prev":null,"subject":"bbcc6c10-0230-43fd-85c8-34d0726851bf","subject_kind":"endpoint"}

Proof

Fold these into the leaf in order, hashing each sorted pair with keccak256, to reach the root of (273461 rows).

  1. 0x27d2e0e71372137018fe93bc216af44bf723298c3e40dca7a1b6ddf7911d5e2b
  2. 0x568d09d65e27c95fbddb1043a35e5dc626adbe2ea078f7154937b0e438a457f7
  3. 0x5e0515bf0352fbf61a7b80998c2a9ff66882f73dee907c4e8c0a4c2eb6992250
  4. 0xd74ed70ab46ade5f53b9f225dfd196bd987ead666fba250699baf3ad9e1c50cb
  5. 0x14f17b21830d92598b4af7fbc562d231ccc93426bdf903e8d5330682efe94749
  6. 0xa44877cd24beca9f2239c74c12b98d28c66c0d0af7553c516748c190546dc931
  7. 0x36b339da11a140a26480d3a58b9bfb23850d3378bfefed8832b3afdf5a605d7e
  8. 0xccd8f3ed6e649c388c9d3f034067d7ce1cd0205599dca04983f6a7af27cd8c1d
  9. 0xbbb92828100c98f02b6f3f9c1e0ee00ba18517427a5726fcac15c23e39a8e1ed
  10. 0x58a55125c2389f16bd21647419b9fa69b354f91d8dd589440e8095b1281b3947
  11. 0xbae4a65d8ae80540633da98a42bc7238b1aae389589d7285d2f85046a743e3a5
  12. 0xf5aa0adf486bca586314afc4935e732d521d3d8275e51eb4c8d2de46b00ea2c1
  13. 0x28be2f2303fcac4decdb6323b0f4ee8f8a31038beb758eaafc604a98626292e4
  14. 0x8ff18b1523f9fbdd16a5796cbe184b46854c06256508af8de5509b598370c725
  15. 0xa383a1cfe9b3f6af67eb4c7642564b9424d46d7fca90e7ba175d549f8b67dd0c
  16. 0x52efa60e0c3d8525486d2310eb3b6b566c9677c64fd1f819edb072cf101bcbdb
  17. 0x22c1fb9487ee0b9732fa017a0cc321eb27a60c08450d112f755089796e12cb1c
  18. 0xa20d48798a34aba12eb9c7297722982d910f06c148f5e27ac59307422964a585
  19. 0x9fc354c0ea3efc3f1b6548bb77ea902a50f2c9647647ea61592efce2bd795699
Day root
0xdd77e5e99da32c3f154180d63e3c3e6fe11e896a2c6896412ef6b5c4bcf7e795
Chained
0x23bb4dfd09f55c15ac3962f53578e02c6114d5bfd9bb451cf8794fb6e2817ef4
Published
On Arc Testnet, block 69414147: 0xb2ebd83aafeb765a0ce66e76aefb133ed06606c98038fd56191e2bcd1dcb77b4

How to check it · This row as JSON