|
Veval — LLM Regression Testing
Veval captures real production traces and turns them into a regression baseline. When your prompt changes, know what broke before your users do.
|
78 % |
|
Perfall | Stop benchmark regressions before merge
|
78 % |
|
docpipeline
|
77 % |
|
Tapedeck · deterministic backtesting for TypeScript
An event-driven backtesting and paper-trading engine for TypeScript. No lookahead, fixed-point money, and it reports what it could not know. Run one free in...
|
77 % |
|
Load testing for every narduk app | loadtest.dev
Declare an HTTP load plan for an owned Narduk app, submit it through the authenticated API, and inspect latency, error rates and threshold results.
|
77 % |
|
Frontguard — Catch the regression, not the noise
AI-powered visual regression testing. AI vision tells a real regression from an intentional change or content, so a red run means something again. Open-sourc...
|
77 % |
|
SeekSpeed Terminal — Make your agents faster, honestly.
A performance terminal for AI agents. Benchmark latency, find bottlenecks, and optimize real agent performance across prompts, tools, and model backends.
|
77 % |
|
stock87 — you found the right what. we tell you about the when.
stock87 — a deterministic timing-verdict API for quant desks. POST the shortlist your models already chose; get back, per symbol, whether now is the moment t...
|
77 % |
|
ffsdb - Rust graph, vector, and writeback database engine
FFS is a Rust database engine for graph, vector, and writeback workloads at high throughput, built for app and agent systems that need retrieval, traversal,...
|
76 % |
|
CompileBench
Benchmark of LLMs on real open-source projects against dependency hell, legacy toolchains, and complex build systems.
|
76 % |
|
Latencio — Load Test Analysis & Performance Verdict Tool
Upload a JMeter JTL, k6, or Gatling result and get an engineering-grade verdict in under 60 seconds. Latencio correlates load test data with Prometheus, APM,...
|
76 % |
|
LINC Forge — every run published
The developer site of LINC Innovations: a pre-registered benchmark of four agent-operated mobile tool stacks, ours included, with every run published.
|
76 % |
|
Best LLM Right Now: AI Model Rankings by Public Benchmarks | BasedAGI
The best LLM overall and for coding, writing, and math, ranked from public benchmark results with source dates.
|
76 % |
|
Vizra Evals — Agent Evals for the Laravel AI SDK, Written in Pest
Built for the official Laravel AI SDK: write agent evals as Pest tests, record every run — scores, judge reasoning, cost — and fail CI on regressions instead...
|
76 % |
|
OpenPriors — the judgement ledger
Models can reliably judge subtle properties of text. OpenPriors harnesses that into a public ledger: cardinal scores with uncertainty, fitted from pairwise r...
|
76 % |
|
Loguro | The Log Operating System
The Log OS. Query 50M logs in under 500ms with human-readable syntax. Replay incidents chronologically, detect recurring patterns forever, and act — create J...
|
76 % |
|
AgentBench — Benchmark Your Claude Code Setup
40 tasks across 7 domains with rule-based scoring. Benchmark, compare, and optimize your Claude Code configuration.
|
76 % |
|
|
76 % |
|
WithAgents — Field notes from agentic development
Writing, products, and open source on building production software with AI coding agents. A WithAgents field journal — patterns, tools, and field notes from...
|
76 % |
|
Hayvenhurst: the live code graph that tells your agents what to test, send, and skip
Hayvenhurst keeps a live graph of your codebase and puts it to work: run only the tests a change can break, hand AI agents the exact slice of code they need...
|
76 % |
Detected social media networks