math benchmark
Humanity's Last Exam (with tools, text-only) Leaderboard
Humanity's Last Exam (with tools, text-only) is a text-only variant evaluated with tool use enabled.
Updated Oct 23, 2026
Models10
Model coverage10
MetricScore
EvidenceB
Humanity's Last Exam (with tools, text-only) Ranking
Higher score ranks better on this benchmark.
Sort by
Rank | Model | Score | Percentile | Participants | Evidence | Evaluated |
|---|
| Rank01 | ModelAN | Score67.70% | Percentile100.00% | Participants10 | EvidenceC | Evaluated |
| Rank02 | ModelDE | Score60.00% | Percentile88.89% | Participants10 | EvidenceC | Evaluated |
| Rank03 | ModelOP | Score57.20% | Percentile77.78% | Participants10 | EvidenceC | Evaluated |
| Rank04 | ModelAC | Score56.20% | Percentile66.67% | Participants10 | EvidenceC | Evaluated |
| Rank05 | ModelTE | Score55.40% | Percentile55.56% | Participants10 | EvidenceC | Evaluated |
| Rank06 | ModelTE | Score53.20% | Percentile44.44% | Participants10 | EvidenceC | Evaluated |
| Rank07 | ModelAC | Score49.80% | Percentile33.33% | Participants10 | EvidenceC | Evaluated |
| Rank08 | ModelST | Score47.20% | Percentile22.22% | Participants10 | EvidenceC | Evaluated |
| Rank09 | ModelTM | Score46.00% | Percentile11.11% | Participants10 | EvidenceC | Evaluated |
| Rank10 | ModelBY | Score40.90% | Percentile0.00% | Participants10 | EvidenceC | Evaluated |
Humanity's Last Exam (with tools, text-only) Highlights
The leading models and scores on this benchmark.
Humanity's Last Exam (with tools, text-only) Score Distribution
A closer view of the leading scores on this benchmark.
The Top AI Models for Humanity's Last Exam (with tools, text-only)
The first five results on this benchmark, with official price and output speed added where the model identity can be matched.
What is Humanity's Last Exam (with tools, text-only)?
What Humanity's Last Exam (with tools, text-only) measures and how its scores work.
A text-only variant of Humanity's Last Exam evaluated with tool use enabled.
It reports a Score as a ratio.
Scores are shown in ratio. This benchmark is not independently verified and has an evidence level of B.
- Family
- Humanity's Last Exam (with tools, text-only)
- Modality
- text
- Primary category
- math
- Score direction
- higher
- LLMBoard eligible
- No
- Evaluation key
- overall
LLM Stats. Benchmark scores retain their original unit. Overall score eligibility is shown separately.
FAQ
Common questions about Humanity's Last Exam (with tools, text-only).
Which model scores highest on Humanity's Last Exam (with tools, text-only)?
Claude Opus 5.5 is currently ranked first with 67.70%.
What are the top three models on Humanity's Last Exam (with tools, text-only)?
The current leaders are Claude Opus 5.5 (67.70%), DeepSeek-V4-Pro-0813 (60.00%), and GPT-6 Astra (57.20%).
Which Humanity's Last Exam (with tools, text-only) model has the lowest official input price?
Step 3.7 Flash has the lowest matched official input price at $0.20 input / $1.2 output per 1M tokens.
Which models are fastest among Humanity's Last Exam (with tools, text-only) results?
The fastest matched records are Claude Opus 5.5 (12.33 tok/s via Anthropic).
Does the highest score result prove overall model quality?
No. This benchmark measures one defined capability or task. The overall LLMBoard score uses a separate aggregation across eligible benchmark evidence.