Humanity’s Last Exam paper
Primary paper introducing HLE.
← Back to Compute College
Compute College
What is Humanity’s Last Exam?
Learn what Humanity’s Last Exam measures and why frontier academic benchmarks matter for model capability claims and AI compute demand.
Plain-English definition
Humanity’s Last Exam, or HLE, is a difficult multimodal benchmark created to test frontier model capability across expert-level academic subjects with broad coverage.
Memory trick: Hard test, not a business case by itself.
Why it matters
A frontier capability claim becomes relevant to ComputeTape when it causes buyers to use costly models for research, analysis, coding, or agent workflows that raise inference demand.
- A frontier capability claim matters here only when it changes buyer behavior.
- That means moving research, analysis, coding, or agent work to costly models.
- A record on a hard exam is not, by itself, a production readiness statement.
Simple example
Progress on a demanding academic test can indicate capability improvement, but it does not state the number of tokens, latency, or dollar cost needed to complete a business task.
- Progress on a demanding academic test indicates capability improvement.
- It does not state tokens, latency, or dollars to finish a business task.
- HLE is multimodal and broad, so a single number compresses many distinct skills.
Example figures are illustrative calculations, not current quoted market prices.
Current example
Primary source
The Humanity’s Last Exam paper introduces HLE as a multimodal benchmark at the frontier of human knowledge with broad subject coverage. Last checked: May 24, 2026.
The page explains the test and avoids unsupported frontier-model comparisons.
Market signal
How to read the market signal
When a model improves on demanding benchmarks, observe whether customers route valuable workloads to it and whether that changes model-serving demand.
- When a model improves on hard benchmarks, watch whether customers route valuable workloads to it.
- Routing — not the headline — is what changes model-serving demand.
- Pair any frontier claim with workload-specific quality, latency, and price evidence.
Market read: an HLE gain is a capability claim; it moves the compute market only when buyers route high-value work to the model. Figures here are illustrative unless explicitly sourced and dated — see our methodology.
Common mistake
Do not equate a hard-test result with universal production readiness or make broad capability claims beyond the cited benchmark.
Practical takeaway
What you can do with this
Use HLE as one frontier evidence source and pair it with workload-specific quality, latency, price, and adoption evidence.
- Use HLE as one frontier evidence source, not a buying decision.
- Pair it with workload-specific quality, latency, price, and adoption evidence.
- Avoid extrapolating beyond the cited benchmark.
Decision check: beyond the HLE number, do you have workload-specific quality, latency, price, and adoption evidence for this model?
Compute College
Follow model releases as market signals
Follow model releases as AI compute market signals in the ComputeTape Morning Brief.
Compute College track
Model Benchmarks & AI Compute Economics
Step 18 of 23: What is humanitys last exam