# ATOM Pricing Intelligence

> The Global Price Benchmark for AI Inference. 1,600+ SKUs, 40+ vendors, 14 price indexes.

- name: io.github.A7OM-AI/atom-mcp-server
- version: 1.1.2
- connection class: R0 (autonomous)
- trust: 86/100
- quality: 64/100 (needs work)
- owner: not claimed
- descriptor: https://api.protogrid.dev/v1/servers/io.github.A7OM-AI%2Fatom-mcp-server
- tools: https://api.protogrid.dev/v1/servers/io.github.A7OM-AI%2Fatom-mcp-server/tools

**An agent can connect right now, no human step.** Remote endpoint, no authentication, reachable on the last probe.

## Connection (mcpServers)

```json
{
  "mcpServers": {
    "atom-mcp-server": {
      "type": "sse",
      "url": "https://atom-mcp-server-production.up.railway.app/mcp"
    }
  }
}
```


## Trust

- hygiene: 100
- liveness: 100
- freshness: 45
- provenance: 90
- drivers: +repository +namespace-matches-repo +website +reachable +uptime-100% -stale-227d
- Derived from observable signals (official registry feed and our own credential-free probes); no code audit performed.

## Quality 64/100 (needs work)

### protocol: 50 (weight 25)

- warn `protocol.modern`: newest supported version is 2025-11-25; 2026-07-28 not supported, older versions are deprecated until 2027-07-28
- n/a `protocol.stateless`: only applies to 2026-07-28 servers
- warn `protocol.transport`: only the legacy HTTP+SSE transport, deprecated until 2027-07-28
- n/a `protocol.list_ttl`: only applies to 2026-07-28 servers

### authorization: n/a (weight 15)

- n/a `auth.prm`, `auth.as_metadata`, `auth.cimd`: no remote uses OAuth
- n/a `auth.secret_in_url`: no templated URL

### tool hygiene: 93 (weight 30)

- pass `tools.descriptions`: every tool has a description
- warn `tools.description_length`: 1 tools have descriptions over 1,024 characters, which crowds the context
- pass `tools.schemas`: every tool has an object input schema
- pass `tools.annotations`: 11 of 11 tools declare readOnlyHint or destructiveHint
- pass `tools.directory_hints`: every tool declares readOnlyHint, destructiveHint, idempotentHint and openWorldHint
- pass `tools.token_cost`: about 3,501 tokens to load every tool
- pass `tools.api_dump`: tools are not a one-to-one API dump

### stability: 25 (weight 15)

- warn `stability.changes`: 13 tool changes in 30 days
- fail `stability.rug_pull`: 9 tools kept their name but changed most of their description in 30 days; review before trusting them

### dependencies: 67 (weight 15)

- pass `deps.known_vulns`: no known advisory in 98 resolved packages
- fail `deps.mcp_sdk_version`: @modelcontextprotocol/sdk 1.30.1 has 1 known advisory (GHSA-6qxp-vccf-f47h); update @modelcontextprotocol/sdk to 1.31.0 or later
- pass `deps.resolvable`: atom-mcp-server@1.1.2 resolved: 98 packages

- tools: 11, about 3,501 tokens to load them all
- tool set hash: 4196dc7fd66f24d54482c12ae139b6cc
- tools last changed: 2026-10-05T19:50:30.074Z
- badge: [![protogrid quality](https://protogrid.dev/badge/io.github.A7OM-AI/atom-mcp-server.svg)](https://protogrid.dev/servers/io.github.A7OM-AI/atom-mcp-server)
- Checks run on what our credential-free, read-only probes observe; tools are never called and no code audit is performed.

### Tool changes

- 2026-10-05 `get_price_history` changed (description)
  - before: Weekly history. With index_code: the index series (benchmark level, May 2026 = 100, and spot price) 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
  - after: 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
- 2026-10-05 `get_index_benchmarks` changed (description)
  - before: 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) and coverage counts. 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"
  - after: 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"
- 2026-09-28 `get_vendor_catalog` changed (title, description, inputSchema) **meaning changed**
  - before: Full catalog for a specific vendor: all models, modalities, and pricing. Returns vendor metadata (country, region, pricing page URL) plus every model and SKU they offer. Examples: - "What does Together AI sell?" → vendor="Together AI" - "OpenAI's text model pricing" → vendor="OpenAI", modality="Text" - "Amazon Bedrock catalog" → vendor="Amazon Bedrock"
  - after: 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).
- 2026-09-28 `compare_prices` changed (title, description, inputSchema) **meaning changed**
  - before: Cross-vendor price comparison for a specific model or model family. Shows the same model (or family) priced across different vendors, sorted cheapest first. Essential for cost optimization and vendor selection. Examples: - "Compare Llama 3.1 70B pricing across vendors" → model_name="Llama 3.1 70B" - "Cheapest GPT-4 family output pricing" → model_family="GPT-4", direction="Output" - "Claude pricing comparison" → model_family="Claude"
  - after: The same model, or a whole family, priced across every vendor that sells it: cheapest and dearest 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"
- 2026-09-28 `get_model_detail` changed (title, description, inputSchema) **meaning changed**
  - before: Deep dive on a single AI model: technical specs + pricing across all vendors. Returns model_registry data (context window, parameters, open-source status, training cutoff, model family) plus all SKU pricing across every vendor that offers this model. Examples: - "Tell me everything about GPT-4o" → model_name="GPT-4o" - "Claude Sonnet 4.5 specs and pricing" → model_name="Claude Sonnet 4.5"
  - after: 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.
- 2026-09-28 `search_models` changed (title, description, inputSchema) **meaning changed**
  - before: Search and filter AI inference models across all tracked vendors and SKUs. Query by modality (Text, Image, Audio, Video, Multimodal), vendor, creator, model family, open-source status, price range, context window, and parameter count. Returns matching models with pricing. Free tier shows count + price range; paid tier shows full details. Examples: - "Find open-source text models under $1/M tokens" → open_source=true, modality="Text", max_price=0.001 - "What multimodal models does Google offer?" → vendor="Google", modality="Multimodal" - "Models with 128K+ context window" → min_context_window=128000
  - after: 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"
- 2026-09-28 `list_vendors` changed (title, description, inputSchema) **meaning changed**
  - before: List all AI inference vendors tracked by ATOM. Returns vendor name, country, region, and pricing page URL. Vendors span four channel types: Model Developers, Cloud Marketplaces, Inference Platforms, and Neoclouds. Optionally filter by region or country. Examples: - "List all vendors" → (no params) - "European AI vendors" → region="Europe" - "Chinese AI vendors" → country="China"
  - after: Every vendor in the Attic Standard fleet with its channel (model developer, cloud marketplace, inference platform or neocloud), country, region and pricing page. Filter by channel, region or country.
- 2026-09-28 `get_market_stats` changed (title, description, inputSchema) **meaning changed**
  - before: Aggregate AI inference market intelligence. Returns total vendor/model/SKU counts, price distribution (median, mean, quartiles, min/max), and modality breakdown. Optionally filter by modality. Examples: - "AI inference market overview" → (no params) - "Text model pricing statistics" → modality="Text" - "Image generation market stats" → modality="Image"
  - after: Coverage of the tracked market (active vendors by channel, 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).
- 2026-09-28 `get_index_constituents` added
  - 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"
- 2026-09-28 `get_model_intelligence` changed (title, description, inputSchema) **meaning changed**
  - before: ATOM Model Intelligence — 6 capability and coverage metrics derived from the metadata behind every tracked model. Complements the pricing KPIs in get_kpis. Returns 6 metrics: - Reasoning Tier Share: % of general-purpose text models that are reasoning-tier - Long-Context Saturation: % of models shipping 128K+ context windows - Frontier Context Ceiling: context multiplier between top-decile and median models - Output Ceiling Spread: max output token multiplier between top-decile and median - Training Cutoff Lag: median months between model training cutoff and today - Vendor Modality Breadth: median number of modalities offered per vendor Read alongside pricing, these explain why a model is priced the way it is. Available to all tiers. Examples: - "How stale are AI models on average?" → Training Cutoff Lag - "What share of models support long context?" → Long-Context Saturation - "How rare are reasoning models?" → Reasoning Tier Share
  - after: 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.

Full history: https://api.protogrid.dev/v1/servers/io.github.A7OM-AI%2Fatom-mcp-server/changes

## Dependencies

- npm atom-mcp-server@1.1.2: resolved, 97 dependencies, MCP SDK @modelcontextprotocol/sdk 1.30.1
  - GHSA-6qxp-vccf-f47h (high) in `` @modelcontextprotocol/sdk@1.30.1 ``, direct, fixed in `` 1.31.0 ``, advisory summary: "MCP TypeScript SDK: OAuth client could send credentials to an authorization server chosen by the MCP server" https://osv.dev/vulnerability/GHSA-6qxp-vccf-f47h

Dependency graphs from deps.dev; advisories from OSV.dev, including the GitHub Advisory Database and the PyPI Advisory Database; all CC BY 4.0.

## Remotes

- https://atom-mcp-server-production.up.railway.app/mcp (sse, auth none, reachable true, uptime30d 1)

## Packages

- npm atom-mcp-server@1.1.2

## Tools (11)

- `compare_prices`: The same model, or a whole family, priced across every vendor that sells it: cheapest and dearest 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"
- `get_index_benchmarks`: 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"
- `get_index_constituents`: 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"
- `get_kpis`: 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.
- `get_market_stats`: Coverage of the tracked market (active vendors by channel, 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).
- `get_model_detail`: 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.
- `get_model_intelligence`: 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.
- `get_price_history`: 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
- `get_vendor_catalog`: 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).
- `list_vendors`: Every vendor in the Attic Standard fleet with its channel (model developer, cloud marketplace, inference platform or neocloud), country, region and pricing page. Filter by channel, region or country.
- `search_models`: 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"