Uptime (30d)%100.0

Latency (p50)523 msfaster than 36% of servers

Protocol eradual

Tools5

AI Model Radar

Description

Which open models fit your GPU or Mac, measured. Model momentum, GPU rental prices, weekly pick.

What our measurements say

AI Model Radar is a remote MCP server published at aimodelradar.dev. It has been probed 6 times since 9/12/2026. It answered in 6 of them (100.0%), a near-uninterrupted record. Median response time is 523 ms, a delay an agent will notice. It offers a narrow, focused set of 5 tools. On the protocol side it speaks the 2026-07-28 stateless spec.

Dual-era2026-07-28 ✓Website ↗Version: 1.1.0First probe: 2026-09-12 15:21:42Last probe: 2026-09-13 01:33:33

Agent readability score

Can an LLM agent pick the right tool here — names, descriptions and parameter clarity are assessed.

Agent readability score

Tool names are descriptive but some parameters lack clarity; descriptions are detailed but could improve error handling explanations.based on measured tool list

Strengths

Tool names clearly indicate their purpose (e.g., 'find_fit' for hardware compatibility).
Descriptions provide context for when to use tools, like 'find_fit' for hardware-specific model recommendations.

Weaknesses

Parameter 'kv' in 'find_fit' is cryptic without explanation.
Descriptions for 'gpu_prices' and 'weekly_pick' lack explicit error handling or constraint details.
Tool 'weekly_pick' has a truncated description, reducing clarity.

find_fit — Parameter 'kv' lacks clear definition or context.
weekly_pick — Description is cut off, omitting critical details about the selection process.

Verdict

The tools are generally well-structured with clear purposes, but ambiguous parameters and incomplete documentation hinder usability. Improving parameter explanations and error handling details would enhance reliability.

9/12/2026, 9:32:00 PM

Event narrative

Raw observations

Supported versions

2026-07-28

as declared by server/discover + initialize

Capabilities

tools

Schema score

100

5 tools

Schema compliance

tools/list structure, inputSchema validity, and a functional smoke test — the components of the 0-100 score.

100/100

Excellent5 tools

tools/list schema valid
inputSchema is valid JSON Schema
Smoke test not run this round
Cold-start: cold 67ms → warm 71ms (penalty 0ms)

Tools

Tools the server advertised in the latest measurement — measured, not catalog-claimed.

trending_models2 params
Tracked AI models ranked by Heat Score (0–100, weighted percentiles of measured Hugging Face/OpenRouter signals) with the raw signals, local-run facts (GGUF size, quantization) and links. Models still collecting a week of history have heat=null and rank after scored ones.
- limitinteger
- slugstring
find_fit4 params · 1 required
Which tracked models run on a given GPU or Mac: measured GGUF weights + computed context cache + runtime overhead versus usable memory. Returns the best recommendation and every verdict (EXCELLENT/GOOD/TIGHT/OFFLOAD_REQUIRED/NOT_RECOMMENDED/UNKNOWN) with plain-language reasons. Get hardware ids from list_hardware.
- hardwarestringrequired
- modelstring
- contextinteger
- kvstring
gpu_prices0 params
Median verified on-demand rental price per GPU class on Vast.ai (USD per hour), with min/p75 and offer counts, plus the collection timestamp.
list_hardware0 params
Curated GPU and Mac profiles the fit engine knows — ids, memory, usable memory after margins, bandwidth. Use an id with find_fit.
weekly_pick1 param
The current pick of the week: one tracked model chosen by a published rule (largest counted Heat Score rise among models that run comfortably on a consumer card of up to 24 GB), with the numbers frozen at selection time, a device-by-device fit ladder and the written report including its caveats. Pass week (e.g. 2026-w37) for a past issue. issue is null until the first issue is published.
- weekstring

Availability history

Event log

Derived by comparing consecutive probes — changes in era, protocol version, build and reachability.

12 hours agoAdded to watch — era: Dual-era

Get your badge

Add this badge to your README — it updates automatically as measurements change.

Markdown

[![mcpmetrics](https://mcpmetrics.io/badge/dev.aimodelradar/ai-model-radar/era.svg)](https://mcpmetrics.io/servers/dev-aimodelradar-ai-model-radar)

HTML

<a href="https://mcpmetrics.io/servers/dev-aimodelradar-ai-model-radar"><img src="https://mcpmetrics.io/badge/dev.aimodelradar/ai-model-radar/era.svg" alt="mcpmetrics"></a>

Raw observations

You are seeing the last 7 days. Sign up for the full history. Which check failed and why is in the dashboard.

Similar MCP servers

The catalog entries whose name and description are closest to this one, found with the same index the search box uses.

Hotel & Vacation Rental MCP - live prices & reviews across Booking, Airbnb, Vrbo & Google Hotelsmcp.stayingapi.com
Search hotels & rentals: live prices & reviews across Booking.com, Airbnb, Vrbo & Google Hotels.
Kaneme — Your writing identity, measured and portablekaneme.com
Build a voice from your own texts, measure whether a text still sounds like you, then write in it.
//beforeyouship — LLM cost modeling from your editorbeforeyouship.dev
Model the realistic monthly cost of an LLM app **before you build it**. Not a token calculator: retries, prompt caching, batch discounts, infra overhead, and 3×/10× growth are modeled in, across GPT-5.x, Claude, Gemini, DeepSeek, and more. **Works without a key.** Connect and ask — demo mode covers the six free-tier models. A Pro API key ([beforeyouship.dev](https://beforeyouship.dev)) unlocks the full 18-model catalog. ## Tools | Tool | What it does | |---|---| - **`estimate_cost`** Full cost model for an architecture at a given usage level. Returns Naive / Realistic / Worst Case $/mo per model, growth scenarios, and an opinionated recommendation. | - **`get_model_prices`** Current per-1M-token pricing (input, output, cached, batch) with context windows and staleness metadata. | - **`list_archetypes`** Seven preset architecture patterns (chatbot, RAG pipeline, multi-step agent, …) used as starting points for estimates. | ## Try it Paste into Claude Code or Cursor after connecting
Open Campus Advisorapi.opencampusadvisor.org
Live academic data across 101 US colleges and universities — directly inside Claude. Search real course catalogs updated every 60 minutes, find faculty with active NIH grants, compare schools head-to-head on any topic, navigate from a career goal to the exact courses and professors that get you there. 19 tools covering course search, faculty research, degree requirements, career outcomes, and cross-school comparison. Set your student profile once and every subsequent call filters and ranks results automatically completed courses removed, results ranked by your career targets and interests. Used for: choosing between schools, planning a semester, finding a research advisor, verifying prerequisites, mapping a path from "I want to work in climate policy" to the specific courses and funded labs at Yale, Columbia, and MIT that lead there.

Comments

No comments yet. Be the first.

https://aimodelradar.dev/api/mcp

2026-09-13 01:33:33	Dual-era	200	676	200	788	2026-07-28
2026-09-12 23:31:36	Dual-era	200	523	200	464	2026-07-28
2026-09-12 21:29:15	Dual-era	200	952	200	1016	2026-07-28
2026-09-12 19:27:29	Dual-era	200	385	200	310	2026-07-28
2026-09-12 17:24:09	Dual-era	200	345	200	278	2026-07-28
2026-09-12 15:21:42	Dual-era	200	468	200	796	2026-07-28

AI Model Radar

Description

What our measurements say

Agent readability score

Agent readability score

Strengths

Weaknesses

Verdict

Event narrative

Raw observations

Schema compliance

Tools

Availability history

Event log

Get your badge

Similar MCP servers

Comments

Report this server

Era history

Response time history