Models & Frontiers
What the new models can actually do, how they were trained, and whether the benchmarks mean anything. Open source vs closed, and where the research is heading.
Guides and explainers
Detailed guides and practical technical analysis.
Latest analysis
Recent research, benchmark reviews and technical updates.
Practical tools
Templates for budgets and project planning
BoredTools offers practical spreadsheets for budgets, freelance work and small projects.
Coding Agent Benchmarks Hit the Generalization Wall
Scale's SWE-Bench Pro public leaderboard reports that top models scoring above 70% on SWE-Bench Verified fall to 23.3% for OpenAI GPT-5 and 23.1% for...
Agent Test-Time Scaling Needs Reuse, Not More Rollouts
General AgentBench reports that running agents for more interaction steps or more sampled trajectories did not reliably improve ten leading agents,...
Coding Agents Need Trajectory Reviews, Not Pass Bits
Most coding-agent benchmarks still compress a whole run into one bit: did the task pass? AgentLens argues that users experience the whole trajectory...
Inference Optimization: From 10x Cost to 10x Speed
In late 2022, running a query against GPT-3-class performance cost roughly $20 per million tokens. By March 2026, multiple models exceed that same...
Scaling Laws Explained for Practitioners: What Actually Matters in 2026
Scaling laws promised a simple deal: spend more compute, get better models. For three years, that deal held. Kaplan et al. drew the first power-law curves...
Multimodal Agents Are Still Missing the Workflow
Multimodal agents can see and act in interfaces, but production value still depends on workflow grounding, reliable UI actions and verification.
Small-Model Routing With Frontier Fallback: The Production Cost Pattern
Small-model routing cuts inference bills only when fallback is measured, budgeted and guarded against confidence failure.
Browser-Use Agents After the Computer-Use Benchmarks
Browser-use agents look cleaner than desktop agents, but the benchmarks still hide drift, cost, auth, and recovery failure.
Models Training Models: The Promise and Peril of Synthetic Data
Microsoft's Phi-4 trained on more than 50% synthetic data and beat GPT-4o on graduate science benchmarks. The old rules about training data are changing fast.
Small Agent Models Need Tool Floors, Not Parameter Claims
Small language models are getting a serious agent story, but the useful question is no longer whether a 1B, 3B, or 8B model can sound capable. The useful...