
Tokenstead - Find AI Models for Your Hardware
Find AI models optimized for your hardware with performance and pricing estimates.
cdnsteve · HN
Tokenstead, find AI models for your hardware
The full gallery
Tech stack
60 projects

Find AI models optimized for your hardware with performance and pricing estimates.
cdnsteve · HN
Tokenstead, find AI models for your hardware

Route LLM calls to cost-effective models without sacrificing quality.
george_avila · Product Hunt
IQ Routing Trajectory-aware LLM routing that cuts agent cost

Grades AI agents' real conversations with an LLM judge, providing A–F scorecards and FinOps analysis.
@tech_maju · X

Monitor AI costs by feature, forecast spending, and set guardrails for your team.
@eastbase_studio · X

Access 200+ AI models (Claude, GPT, Gemini) through one unified API with automatic failover.
@MixRoute_ai · X

Detect AI/LLM generated code in git repositories via commit history analysis.
ava · Lobsters

Monitor AI API costs from OpenAI, Anthropic, and xAI by connecting provider keys.
@ylin_ · X
So my launch of – the simple, proxy-less AI model cost monitoring software – didn’t generate much response. I said I would build this in public, so here is an update. I am very happy with the app for my own use. It’s everything I want it to be. So even without users, I’m glad I built this just for myself. It’s probably not accurate to say I “launched” this. All I did was publish the app, make a single post on X, LinkedIn and Show HN. That’s not really any sort of distribution effort needed for an app. So I will continue to get the word out regarding Costbase on Reddit, and other places, as well as engaging with X users who have problem with AI cost. (If you guys have ideas, lmk) It’s just a small app, a side project, so I won’t put too much time marketing this before calling it quit. Pre-AI, I’ve always spent so much effort building an app that I don’t have any time left for marketing (I also run an ERP consultancy). Now, building is pretty much automated, so th

Scores AI-generated ad creative and returns verdicts (run/fix/kill) via MCP and REST API.
ds246 · HN
Spendict – a performance marketer's verdict for AI agents, over MCP

See where your AI spending drives value and the highest-ROI improvement areas.
@pgrant67 · X
Uber gave 5,000 engineers Claude Code. 4 months later, their entire 2026 AI budget was gone. This is what happens when adoption outruns proof. We built the free AI ROI Index to show where your AI spend is actually driving value:

Visualize hardware performance metrics while running LLM inference on your system.
dev_dan_2 · HN
WatchMachineGo – A visualizer to show hardware performing LLM inference

Check OpenAI-compatible API endpoints for model quality issues, degradation, and wrapping risks.
AI快站模型质量检测 — 面向 OpenAI Compatible 接口的网页检测工具,输入公开 HTTPS 地址和临时 API Key,可检查模型声明、Token、动态题、SSE 与工具调用并生成分项报告;密钥仅用于当次检测,不写入数据库、缓存或日志

Analyzes code diffs and changelogs using AI with your own API keys.
@eldrex_bula · X