Evals
How models and agents are measured: benchmarks, leaderboards, evaluation suites and hallucination rates
Latest stories See all →
- Top 50 AI & LLM Evaluation Interview Questions and Answers for 2027
- The 99% Accuracy Trap: Why Is It Failing in Production?
- Nemotron 3.5 Bolt: Specs, Benchmarks and Size (2026)
- How to Migrate SharePoint RAG to a Production-Ready Pipeline with Azure AI Search and Jinba
- Particle-1 Benchmark Results for Data Analysis
- LLM as a Judge: How to Use One Model to Evaluate Another Without Fooling Yourself
- My Multi-Agentic framework scored 100% on Accuracy Task Completion — It Was Not the Most Important…
- AI Benchmarks Can Change Without a Single Code Commit
- Evaluation Metrics in Machine Learning: Classification vs. Regression
- AI Fluency Benchmarks: What Good Looks Like by Role
- GPT-6.1 Terra vs. GPT-6 Astra: One sacrifices accuracy for lower cost and higher speed at high effort
- Evil Qwen and activation steering: why SBOMs aren’t enough
- GPT-6.1 Terra Draws a Python Reading a Book
- From synthetic enterprise data evaluations to real Microsoft Copilot AI confidence
- Does ‘eval’ in the workspace path change what your agent does?
- How Deterministic AI Produces Predictable, Auditable Outputs
- Fine-Tuning Is Super, Super Hard. Even OpenAI Is Backing Away From It.
- Adding an AI agent to a Webflow site, and what it does to your speed scores
- Samaya AI Releases Sterling: A 3B-Active Finance Research Agent Model that Beats Claude Fable 5.1 on FrontierFinance at 11x Lower Cost
- How to Evaluate a Machine Learning Model: Accuracy, Precision, Recall, and More
- Nanbeige4.3-33B: Benchmarks for the New Open Agentic Coding Model
- How You Test an Agent That Runs for 30 Days Before It Touches a Live Loan: Golden Files, Backtesting, and the Eval Harness SR 11-7 Actually Wants
- Cohere Command A+ Benchmark Results for Data Analysis
- 3 ways to improve agents: simulations, traces and evaluations
Videos
- Pre-Push Eval Gate: Block AI Agent Regressions Before They Ship | Splunk Agent Observability
- Autotune: 81% to 96% Eval Accuracy in One Click | Splunk Agent Observability
- Observability and Evals Masterclass
- Gemini 4 Carbon: Frontier Diamond or Just Benchmarks?
- Nanbeige4.3 33B: Great Benchmarks, But It Failed My Tests Royally
- How to Write Your First AI Eval - SallyAnn DeLucia, Arize AI
- Do Not Let Models Invent Numbers - Vinoo Ganesh, Kepler
- AI Avatars IGNORE You?! I Tested the CRAZY New StreamEval Benchmark (RTX 4090)
- I Ran Flint, the CRAZY New 40B Coding Model, WITHOUT a Mac. Did It Fix My Buggy Repo?
- Don't Trust Your AI Agents. Test Them Like a Scientist — Stephanie Jarmak, Omni Analytics
- Why AI Writing Still Reads Like Slop — Louis-François Bouchard, Towards AI
- Agentic AI System Design: Evals, Guardrails and Cost
Podcasts
Recent episodes
- Hackers leaked details of more than 41,000 Homeland Security personnel, the FCC wants to let prisons jam illegal cellphones, and Google's Gemini 4 Carbon promises 'significant leap' in AI coding and cybersecurity
- Artificial Analysis v5 and the benchmarker's dilemma
- SaaStr 884: The AI-First Seller: How Databricks Rebuilt Its Sales DNA at $190B with CRO Ron Gabrisko
- ⚡ An Eval for AI Scientists That Knows the Answer Key - Ian Quigley, Leash Bio
- Podcast: Kimi K3.1 Pro, the new cyber reality, and Muse musings
- AI Agents Need a Babysitter
- Can data agents do real analysis?
- What I learned from cheating clankers
- A New Way to Measure AI Safety for Kids
- Rewriting History with AI
- Eric Schmidt's $80 Billion Anthropic Bet
- Meta and Google Rejoin the Frontier, Microsoft's Maia 300, and the AI Credit Scare | The Six Five Pod Ep. 324