Attic Standard

Measured conclusion: Alive, not deep-tested. L2 outcome: not_attempted.

The global price benchmark for AI inference: 23 price indexes, market KPIs and SKU-level prices.

Add to your agent

Copy this into your MCP client config:

{
  "mcpServers": {
    "Attic Standard": {
      "url": "https://mcp.atticstandard.com/mcp",
      "transport": "streamable-http"
    }
  }
}

Measured tools (11)

• get_index_benchmarks read-only

  The Attic Standard price indexes for AI inference: every published index at the week the site shows. Six families: Modality (text, multimodal, image, video, audio, voice, embeddings), Channel (model developers, cloud marketplaces, inference platforms, neoclouds), Tier (flagship, core, compact), License (open weights, restricted weights, proprietary), Origin (United States, China) and Use case (reasoning, coding). For each index and direction (input, cached input, output) it returns the benchmark level (May 2026 = 100) with week, month and vs-base changes, the spot price (median, p25, p75 in dollars) with its own week, month and vs-base changes, and coverage counts with the May 2026 basket size, flagged where the basket has changed materially since the base. Free: fully public, the same figures atticstandard.com publishes. Token prices are per 1,000 tokens; other modalities use their own unit (per image, per second, per minute, per 1,000 characters). Examples: - "Where is text inference priced this week?" -> index_code="TXT" - "Compare the four channels" -> index_category="Channel" - "Open-weight versus proprietary" -> index_category="License"

  Parameters:

  • index_code — string — One index, e.g. 'TXT', 'NCL', 'FLG' or the full code 'AIPI TXT GLB'. Omit for all published indexes.
  • _atom_api_key — string — Your Attic Standard MCP PRO key for vendor- and SKU-level data. Omit for the free tier.
  • index_category — string — Index family: 'Modality', 'Channel', 'Tier', 'License', 'Origin', 'Use case'.

  Hints: read-only yes · destructive no · idempotent yes · open-world no

• get_price_history read-only

  Weekly history. With index_code: the index series (benchmark level, May 2026 = 100, and the spot price with its 25th and 75th percentiles) for up to 260 weeks; free. With model_name: every vendor's price for that model week by week, showing when each SKU was repriced; Attic Standard MCP PRO. Token prices are per 1,000 tokens; other modalities use their own unit (per image, per second, per minute, per 1,000 characters). Examples: - "How has the neocloud index moved since May?" -> index_code="NCL" - "Has anyone cut the price of DeepSeek V3 this quarter?" -> model_name="DeepSeek V3", weeks=13

  Parameters:

  • weeks — integer — How many weeks back (default 26)
  • vendor — string — With model_name: limit to one vendor. PRO.
  • direction — string — Pricing direction
  • index_code — string — Index history, e.g. 'TXT', 'DEV', 'OSS' or 'AIPI TXT GLB'. Free.
  • model_name — string — Model price history across every vendor that sells it, e.g. 'GPT-4o'. PRO.
  • _atom_api_key — string — Your Attic Standard MCP PRO key for vendor- and SKU-level data. Omit for the free tier.

  Hints: read-only yes · destructive no · idempotent yes · open-world no

• get_kpis read-only

  The nine market KPIs Attic Standard publishes each week. - Price structure: output premium, caching discount, caching availability - Price dynamics: repricing activity, repricing depth, post-launch drift - Competition: multi-vendor spread, first-party premium, marketplace premium Each KPI returns its value, interquartile range, population, week, month and vs-base changes, and its published definition. Free: fully public, the same figures atticstandard.com publishes.

  Parameters:

  • group — string — Optional group: 'Price structure', 'Price dynamics' or 'Competition'
  • weeks — integer — Weeks of history per KPI (default 1, latest only)
  • _atom_api_key — string — Your Attic Standard MCP PRO key for vendor- and SKU-level data. Omit for the free tier.

  Hints: read-only yes · destructive no · idempotent yes · open-world no

• get_model_intelligence read-only

  Six capability and coverage measures drawn from the metadata behind every tracked model: reasoning tier share, long-context saturation, frontier context ceiling, output ceiling spread, training cutoff lag and vendor modality breadth. Read alongside the market KPIs, they explain why models are priced the way they are. Free: fully public, the same figures atticstandard.com publishes.

  Parameters:

  • _atom_api_key — string — Your Attic Standard MCP PRO key for vendor- and SKU-level data. Omit for the free tier.

  Hints: read-only yes · destructive no · idempotent yes · open-world no

• get_index_constituents read-only

  What sits inside an index basket this week. Free tier: the composition (SKUs, models and vendors, split by channel, origin, tier and license). Attic Standard MCP PRO: the model-by-model basket with the vendors selling each model. Examples: - "What is in the flagship index?" -> index_code="FLG" - "How much of the China index is sold through neoclouds?" -> index_code="CHN"

  Parameters:

  • index_code (required) — string — Index to open, e.g. 'TXT', 'FLG', 'CHN' or 'AIPI TXT GLB'
  • direction — string — Limit to one pricing direction
  • _atom_api_key — string — Your Attic Standard MCP PRO key for vendor- and SKU-level data. Omit for the free tier.

  Hints: read-only yes · destructive no · idempotent yes · open-world no

• get_market_stats read-only

  Coverage of the tracked market (vendors and pricing services, in total, by channel and in published indexes; countries, models, priced SKUs, published indexes) and price distributions split by modality, unit and direction. Free tier returns counts, ranges and redacted samples; Attic Standard MCP PRO ($500/month, https://atticstandard.com/mcp) returns vendor names, model names and exact prices. Token prices are per 1,000 tokens; other modalities use their own unit (per image, per second, per minute, per 1,000 characters).

  Parameters:

  • modality — string — Optional modality: Text, Multimodal, Image, Video, Audio, Voice
  • _atom_api_key — string — Your Attic Standard MCP PRO key for vendor- and SKU-level data. Omit for the free tier.

  Hints: read-only yes · destructive no · idempotent yes · open-world no

• list_vendors read-only

  Every pricing service in the Attic Standard fleet, with its vendor (the company, in parent), channel (model developer, cloud marketplace, inference platform or neocloud), country, region, currency and pricing page. Returns the vendor and pricing-service totals. Filter by channel, region or country.

  Parameters:

  • region — string — Optional region, e.g. 'North America', 'Europe', 'Asia'
  • channel — string — Optional channel: 'Model developer' (DEV), 'Cloud marketplace' (CLD), 'Inference platform' (PLT), 'Neocloud' (NCL)
  • country — string — Optional country, e.g. 'United States', 'China', 'France'
  • _atom_api_key — string — Your Attic Standard MCP PRO key for vendor- and SKU-level data. Omit for the free tier.

  Hints: read-only yes · destructive no · idempotent yes · open-world no

• search_models read-only

  Search every priced SKU by modality, vendor, channel, creator, family, tier, license, origin, reasoning, open weights, direction, maximum price and minimum context window. Cheapest first. Free tier returns counts, ranges and redacted samples; Attic Standard MCP PRO ($500/month, https://atticstandard.com/mcp) returns vendor names, model names and exact prices. Token prices are per 1,000 tokens; other modalities use their own unit (per image, per second, per minute, per 1,000 characters). Examples: - "Open-weight text models under $0.0005 per 1,000 output tokens" -> open_source="true", modality="Text", direction="Output", max_price=0.0005 - "Chinese reasoning models on neoclouds" -> origin="China", reasoning="true", channel="Neocloud"

  Parameters:

  • tier — string — Lineup tier: 'Flagship', 'Core' or 'Compact'
  • limit — integer — Results to return (default 20)
  • offset — integer — Offset for paging
  • origin — string — Creator's home country, e.g. 'United States', 'China'
  • vendor — string — Vendor name or id
  • channel — string — 'Model developer', 'Cloud marketplace', 'Inference platform' or 'Neocloud'
  • creator — string — Model creator (the lab that built the model)
  • license — string — License class, e.g. 'open', 'restricted', 'proprietary'
  • modality — string — Text, Multimodal, Image, Video, Audio, Voice
  • direction — string — Pricing direction
  • max_price — number — Maximum price in the SKU's own unit (per 1,000 tokens for token models)
  • reasoning — string — 'true' for reasoning models only, 'false' to exclude them
  • open_source — string — 'true' for models with published weights
  • model_family — string — Model family, e.g. 'Llama', 'Qwen', 'Claude'
  • _atom_api_key — string — Your Attic Standard MCP PRO key for vendor- and SKU-level data. Omit for the free tier.
  • min_context_window — integer — Minimum context window in tokens

  Hints: read-only yes · destructive no · idempotent yes · open-world no

• get_model_detail read-only

  One model in depth: specs (creator, origin, tier, license, context window, output limit, training cutoff, modalities), the Attic Standard indexes it belongs to, and its price at every vendor. Free tier returns counts, ranges and redacted samples; Attic Standard MCP PRO ($500/month, https://atticstandard.com/mcp) returns vendor names, model names and exact prices.

  Parameters:

  • model_name (required) — string — Model to look up, e.g. 'GPT-4o', 'Claude Sonnet 4.5', 'Llama 3.3 70B'
  • _atom_api_key — string — Your Attic Standard MCP PRO key for vendor- and SKU-level data. Omit for the free tier.

  Hints: read-only yes · destructive no · idempotent yes · open-world no

• compare_prices read-only

  The same model, or a whole family, priced across every vendor that sells it: cheapest and highest offer, vendor count and spread ratio for each direction, plus each offer with its channel. Free tier returns counts, ranges and redacted samples; Attic Standard MCP PRO ($500/month, https://atticstandard.com/mcp) returns vendor names, model names and exact prices. Token prices are per 1,000 tokens; other modalities use their own unit (per image, per second, per minute, per 1,000 characters). Examples: - "Cheapest place to run Llama 3.3 70B" -> model_name="Llama 3.3 70B" - "Qwen family output prices" -> model_family="Qwen", direction="Output"

  Parameters:

  • limit — integer — Maximum SKUs (default 50)
  • direction — string — Pricing direction
  • model_name — string — Model to compare, e.g. 'GPT-4o', 'Llama 3.3 70B', 'DeepSeek V3'
  • model_family — string — Whole family instead of one model, e.g. 'Llama', 'Qwen'
  • _atom_api_key — string — Your Attic Standard MCP PRO key for vendor- and SKU-level data. Omit for the free tier.

  Hints: read-only yes · destructive no · idempotent yes · open-world no

• get_vendor_catalog read-only

  Everything one vendor sells: channel, country, pricing page, model and SKU counts, and the full price list. Free tier returns counts, ranges and redacted samples; Attic Standard MCP PRO ($500/month, https://atticstandard.com/mcp) returns vendor names, model names and exact prices. Token prices are per 1,000 tokens; other modalities use their own unit (per image, per second, per minute, per 1,000 characters).

  Parameters:

  • vendor (required) — string — Vendor name or id, as listed by list_vendors
  • limit — integer — Maximum SKUs (default 100)
  • modality — string — Optional modality: Text, Multimodal, Image, Video, Audio, Voice
  • direction — string — Optional pricing direction
  • _atom_api_key — string — Your Attic Standard MCP PRO key for vendor- and SKU-level data. Omit for the free tier.

  Hints: read-only yes · destructive no · idempotent yes · open-world no

Publisher website Repository

Measured by MCPYP probing. Untested fields are omitted, never estimated. Listing is always free; MCPYP never proxies calls or stores user tokens.