Humanity's Last Exam Diamond without tools at highest reasoning effort (HLE-Diamond (highest effort))
We show this table for reference; we do not rank on it.
Accuracy on the 1,000-question HLE-Diamond set without tools at each model's highest available reasoning effort.
Accuracy without tools at highest reasoning effort on HLE-Diamond (highest effort) — October 5, 2026 source observation
We mirror the published accuracy without tools at highest reasoning effort view for HLE-Diamond (highest effort). GPT-6 Astra leads the public snapshot at 65.8%, followed by GPT-6.1 Sol (61.3%) and Claude Opus 5.5 (60.9%). We do not use these results to rank models overall.
GPT-6 Astra
OpenAI
No tools · max reasoning
GPT-6.1 Sol
OpenAI
No tools · max reasoning
Claude Opus 5.5
Anthropic
No tools · max reasoning
18 modelsKnowledgeCurrentDisplay onlyUpdated October 5, 2026 source observation
Accuracy without tools at highest reasoning effort table (18 models)
| Rank | Model / configuration | Score | Parameters (B) | Open / closed |
|---|---|---|---|---|
| 1 | GPT-6 AstraOpenAINo tools · max reasoning | 65.8% | Not reported | Closed |
| 2 | GPT-6.1 SolOpenAINo tools · max reasoning | 61.3% | Not reported | Closed |
| 3 | Claude Opus 5.5AnthropicNo tools · max reasoning | 60.9% | Not reported | Closed |
| 4 | Gemini 4 ArgonGoogleNo tools · max reasoning | 56.9% | Not reported | Closed |
| 5 | Claude Fable 5.1AnthropicNo tools · max reasoning | 55.1% | Not reported | Closed |
| 6 | Claude Sonnet 5.5AnthropicNo tools · max reasoning | 43.3% | Not reported | Closed |
| 7 | GPT-6 SolOpenAINo tools · max reasoning | 43.2% | Not reported | Closed |
| 8 | GPT-5.6 SolOpenAINo tools · max reasoning | 41.0% | Not reported | Closed |
| 9 | Muse WonderMetaNo tools · max reasoning | 40.6% | Not reported | Closed |
| 10 | Claude Opus 5AnthropicNo tools · max reasoning | 40.5% | Not reported | Closed |
| 11 | Kimi K3.1 ProMoonshot AINo tools · max reasoning | 38.7% | Not reported | Open |
| 12 | Muse Spark 1.3MetaNo tools · max reasoning | 34.6% | Not reported | Closed |
| 13 | Gemini 3.8 FlashGoogleNo tools · high reasoning | 33.3% | Not reported | Closed |
| 14 | Kimi K3Moonshot AINo tools · high reasoning | 28.3% | Not reported | Open |
| 15 | Muse Spark OpenMetaNo tools · max reasoning | 28.2% | Not reported | Open |
| 16 | GLM 5.3Z.AINo tools · max reasoningText-only subset (728 questions) | 25.1% | Not reported | Unknown |
| 17 | Grok 4.7xAINo tools · xhigh reasoning | 25.0% | Not reported | Closed |
| 18 | DeepSeek V4 Pro 0813DeepSeekNo tools · max reasoning | 16.6% | Not reported | Open |
How to read this leaderboard
This view uses the highest reasoning effort available for each listed model on HLE-Diamond, without external tools.
Variant check: The default no-tools table uses high reasoning effort for every model. Its scores and the web-and-code scores have separate tables.
Compare these rows as highest-available-effort system results. An effort setting can change token use, latency, and score, so the numbers should not be mixed with the high-effort table.
Operator receipt: 18 sourced rows are currently displayable on this page; the leading published row is GPT-6 Astra at 65.8%.
Provenance: The scores mirror the reasoning-max view in the Center for AI Safety and Scale AI release. We have not rerun the gated question set.
Source freshness: HLE-Diamond was released on September 22, 2026. The current release table was checked on October 5, 2026; it does not date each score revision.
Honest limit: Highest available effort means different settings across providers. The source does not provide uncertainty intervals. GLM 5.3 reports a text-only subset, and DeepSeek V4 Pro has inconsistent effort labels across its overall score and partition table; neither appears in this full-set mirror.
How to read the highest-effort no-tools table
The HLE-Diamond release has a second no-tools view for each model's highest available reasoning effort. Its 18 scores use the same question sets as the high-effort table. GLM 5.3 ran only the 728 text-only questions (347 reasoning, 381 knowledge), so that score is not a full-set result.
Reasoning effort changes the evaluation setting. We keep this view separate from the default high-effort and web-and-code tables. It is display only and does not move model rankings.
Snapshot
The published HLE-Diamond (highest effort) snapshot places GPT-6 Astra first at 65.8%. The third row is 4.9 points behind. The broader top-10 range is 25.3 points, so the table still separates the published systems.
18 models have been evaluated on HLE-Diamond (highest effort). The benchmark falls in the Knowledge category. HLE-Diamond (highest effort) is currently displayed for reference but excluded from the scoring formula, so it does not directly affect overall rankings.
About HLE-Diamond (highest effort)
Year
2026
Tasks
1,000 expert questions: 500 reasoning and 500 knowledge
Format
Accuracy without tools, highest available reasoning effort
Difficulty
Frontier expert knowledge and reasoning
The release table checked on October 5, 2026 has 12 full-set results at the highest available effort. GPT, Claude, and Muse use max; Gemini and Kimi use high; Grok uses xhigh, as listed by the benchmark owner.
Freshness and provenance
Version
HLE-Diamond, highest available reasoning effort without tools
Refresh cadence
Static
Staleness state
Current
Question availability
Gated 1,000-question dataset; public aggregate results
BenchLM uses freshness metadata to decide whether a benchmark should still be treated as a strong differentiator, a benchmark to watch, or a display-only reference. For the full scoring policy, see the BenchLM methodology page.
Questions
What does HLE-Diamond (highest effort) measure?
Accuracy on the 1,000-question HLE-Diamond set without tools at each model's highest available reasoning effort.
Which model leads the published HLE-Diamond (highest effort) snapshot?
GPT-6 Astra currently leads the published HLE-Diamond (highest effort) snapshot with 65.8% accuracy without tools at highest reasoning effort. BenchLM shows this benchmark for display only and does not use it in overall rankings.
How many models are evaluated on HLE-Diamond (highest effort)?
The October 5, 2026 source observation snapshot contains 18 AI models.
Know when it’s worth switching models
The model to choose, the cheaper alternative, and the release we would wait on.
Read a sample issueJoin 7,000+ subscribers.
One email each week. Unsubscribe anytime.