BENCHMARK EXPLORER

One benchmark. Every result.

Explore rankings, coverage and original sources across 73 registered versions and four existing index snapshots.

PICK A BENCHMARK

CategoryBenchmark and version

Reasoning

AA Intelligence Index

Version snapshot-2026-09-10 (unversioned) · Published board

Primary source ↗

Published AA index from the retained API snapshot. No verified semantic version was supplied.

633 / 835

Catalog configurations with a result

Source identities not yet matched

points

Higher is better · native source scale

Search modelsEvidenceOpen weightsInclude unmatched source identities

633 results match these filters. Coverage above uses the evidence filter before model filters.

AA Intelligence Index, version snapshot-2026-09-10 (unversioned), Published board. Rank is within the filtered results; ties share a rank.
Rank	Model / configuration	Result and provenance	Explore
1	Claude Fable 5.1 (Adaptive Reasoning, Max Effort, Default Fallback) Anthropic · max	53.4 pointsmeasuredobserved 2026-09-10artificialanalysis.ai ↗ Evidence Axis: AA Intelligence Index · snapshot-2026-09-10 (unversioned) · Published board Exact value: `53.4` points Observed: 2026-09-10 · publication date: not recorded Observation id: `legacy:aa_intelligence_index:claude-fable-5.1::max` Source: https://artificialanalysis.ai/leaderboards/models Protocol: inspect · file `data/raw/artificialanalysis.json`	Compare ↗
2	Claude Fable 5.1 (Adaptive Reasoning, Xhigh Effort, Default Fallback) Anthropic · xhigh	53.2 pointsmeasuredobserved 2026-09-10artificialanalysis.ai ↗ Evidence Axis: AA Intelligence Index · snapshot-2026-09-10 (unversioned) · Published board Exact value: `53.2` points Observed: 2026-09-10 · publication date: not recorded Observation id: `legacy:aa_intelligence_index:claude-fable-5.1::xhigh` Source: https://artificialanalysis.ai/leaderboards/models Protocol: inspect · file `data/raw/artificialanalysis.json`	Compare ↗
3	GPT-6 Astra (max) OpenAI · max	52.8 pointsmeasuredobserved 2026-09-10artificialanalysis.ai ↗ Evidence Axis: AA Intelligence Index · snapshot-2026-09-10 (unversioned) · Published board Exact value: `52.8` points Observed: 2026-09-10 · publication date: not recorded Observation id: `legacy:aa_intelligence_index:gpt-6-astra::max` Source: https://artificialanalysis.ai/leaderboards/models Protocol: inspect · file `data/raw/artificialanalysis.json`	Compare ↗
4	GPT-6 Astra (xhigh) OpenAI · xhigh	52.5 pointsmeasuredobserved 2026-09-10artificialanalysis.ai ↗ Evidence Axis: AA Intelligence Index · snapshot-2026-09-10 (unversioned) · Published board Exact value: `52.5` points Observed: 2026-09-10 · publication date: not recorded Observation id: `legacy:aa_intelligence_index:gpt-6-astra::xhigh` Source: https://artificialanalysis.ai/leaderboards/models Protocol: inspect · file `data/raw/artificialanalysis.json`	Compare ↗
5	Claude Fable 5.1 (Adaptive Reasoning, High Effort, Default Fallback) Anthropic · high	51.2 pointsmeasuredobserved 2026-09-10artificialanalysis.ai ↗ Evidence Axis: AA Intelligence Index · snapshot-2026-09-10 (unversioned) · Published board Exact value: `51.2` points Observed: 2026-09-10 · publication date: not recorded Observation id: `legacy:aa_intelligence_index:claude-fable-5.1::high` Source: https://artificialanalysis.ai/leaderboards/models Protocol: inspect · file `data/raw/artificialanalysis.json`	Compare ↗
6	GPT-6 Astra (high) OpenAI · high	51 pointsmeasuredobserved 2026-09-10artificialanalysis.ai ↗ Evidence Axis: AA Intelligence Index · snapshot-2026-09-10 (unversioned) · Published board Exact value: `51` points Observed: 2026-09-10 · publication date: not recorded Observation id: `legacy:aa_intelligence_index:gpt-6-astra::high` Source: https://artificialanalysis.ai/leaderboards/models Protocol: inspect · file `data/raw/artificialanalysis.json`	Compare ↗
7	Claude Opus 5 (Adaptive Reasoning, Max Effort) Anthropic · max	50.7 pointsmeasuredobserved 2026-09-10artificialanalysis.ai ↗ Evidence Axis: AA Intelligence Index · snapshot-2026-09-10 (unversioned) · Published board Exact value: `50.7` points Observed: 2026-09-10 · publication date: not recorded Observation id: `legacy:aa_intelligence_index:claude-opus-5::max` Source: https://artificialanalysis.ai/leaderboards/models Protocol: inspect · file `data/raw/artificialanalysis.json`	Compare ↗
8	Claude Fable 5 (Adaptive Reasoning, Max Effort, Opus 4.8 Fallback) Anthropic · max	49.7 pointsmeasuredobserved 2026-09-10artificialanalysis.ai ↗ Evidence Axis: AA Intelligence Index · snapshot-2026-09-10 (unversioned) · Published board Exact value: `49.7` points Observed: 2026-09-10 · publication date: not recorded Observation id: `legacy:aa_intelligence_index:claude-fable-5::max` Source: https://artificialanalysis.ai/leaderboards/models Protocol: inspect · file `data/raw/artificialanalysis.json`	Compare ↗
8	Claude Opus 5 (Adaptive Reasoning, Xhigh Effort) Anthropic · xhigh	49.7 pointsmeasuredobserved 2026-09-10artificialanalysis.ai ↗ Evidence Axis: AA Intelligence Index · snapshot-2026-09-10 (unversioned) · Published board Exact value: `49.7` points Observed: 2026-09-10 · publication date: not recorded Observation id: `legacy:aa_intelligence_index:claude-opus-5::xhigh` Source: https://artificialanalysis.ai/leaderboards/models Protocol: inspect · file `data/raw/artificialanalysis.json`	Compare ↗
8	GPT-6 Astra (medium) OpenAI · medium	49.7 pointsmeasuredobserved 2026-09-10artificialanalysis.ai ↗ Evidence Axis: AA Intelligence Index · snapshot-2026-09-10 (unversioned) · Published board Exact value: `49.7` points Observed: 2026-09-10 · publication date: not recorded Observation id: `legacy:aa_intelligence_index:gpt-6-astra::medium` Source: https://artificialanalysis.ai/leaderboards/models Protocol: inspect · file `data/raw/artificialanalysis.json`	Compare ↗
11	Claude Fable 5.1 (Adaptive Reasoning, Medium Effort, Default Fallback) Anthropic · medium	49.1 pointsmeasuredobserved 2026-09-10artificialanalysis.ai ↗ Evidence Axis: AA Intelligence Index · snapshot-2026-09-10 (unversioned) · Published board Exact value: `49.1` points Observed: 2026-09-10 · publication date: not recorded Observation id: `legacy:aa_intelligence_index:claude-fable-5.1::medium` Source: https://artificialanalysis.ai/leaderboards/models Protocol: inspect · file `data/raw/artificialanalysis.json`	Compare ↗
12	Claude Opus 5 (Adaptive Reasoning, High Effort) Anthropic · high	48.2 pointsmeasuredobserved 2026-09-10artificialanalysis.ai ↗ Evidence Axis: AA Intelligence Index · snapshot-2026-09-10 (unversioned) · Published board Exact value: `48.2` points Observed: 2026-09-10 · publication date: not recorded Observation id: `legacy:aa_intelligence_index:claude-opus-5::high` Source: https://artificialanalysis.ai/leaderboards/models Protocol: inspect · file `data/raw/artificialanalysis.json`	Compare ↗
12	Muse Spark 1.3 (max) Meta · max	48.2 pointsmeasuredobserved 2026-09-10artificialanalysis.ai ↗ Evidence Axis: AA Intelligence Index · snapshot-2026-09-10 (unversioned) · Published board Exact value: `48.2` points Observed: 2026-09-10 · publication date: not recorded Observation id: `legacy:aa_intelligence_index:muse-spark-1.3::max` Source: https://artificialanalysis.ai/leaderboards/models Protocol: inspect · file `data/raw/artificialanalysis.json`	Compare ↗
14	GPT-5.6 Sol (max) OpenAI · max	47.1 pointsmeasuredobserved 2026-09-10artificialanalysis.ai ↗ Evidence Axis: AA Intelligence Index · snapshot-2026-09-10 (unversioned) · Published board Exact value: `47.1` points Observed: 2026-09-10 · publication date: not recorded Observation id: `legacy:aa_intelligence_index:gpt-5.6-sol::max` Source: https://artificialanalysis.ai/leaderboards/models Protocol: inspect · file `data/raw/artificialanalysis.json`	Compare ↗
15	Claude Fable 5.1 (Adaptive Reasoning, Low Effort, Default Fallback) Anthropic · low	47 pointsmeasuredobserved 2026-09-10artificialanalysis.ai ↗ Evidence Axis: AA Intelligence Index · snapshot-2026-09-10 (unversioned) · Published board Exact value: `47` points Observed: 2026-09-10 · publication date: not recorded Observation id: `legacy:aa_intelligence_index:claude-fable-5.1::low` Source: https://artificialanalysis.ai/leaderboards/models Protocol: inspect · file `data/raw/artificialanalysis.json`	Compare ↗
16	GPT-6 Astra (low) OpenAI · low	46 pointsmeasuredobserved 2026-09-10artificialanalysis.ai ↗ Evidence Axis: AA Intelligence Index · snapshot-2026-09-10 (unversioned) · Published board Exact value: `46` points Observed: 2026-09-10 · publication date: not recorded Observation id: `legacy:aa_intelligence_index:gpt-6-astra::low` Source: https://artificialanalysis.ai/leaderboards/models Protocol: inspect · file `data/raw/artificialanalysis.json`	Compare ↗
17	GPT-6 Astra (Non-reasoning) OpenAI · non-reasoning	45.2 pointsmeasuredobserved 2026-09-10artificialanalysis.ai ↗ Evidence Axis: AA Intelligence Index · snapshot-2026-09-10 (unversioned) · Published board Exact value: `45.2` points Observed: 2026-09-10 · publication date: not recorded Observation id: `legacy:aa_intelligence_index:gpt-6-astra::non-reasoning` Source: https://artificialanalysis.ai/leaderboards/models Protocol: inspect · file `data/raw/artificialanalysis.json`	Compare ↗
17	Muse Spark 1.3 (xhigh) Meta · xhigh	45.2 pointsmeasuredobserved 2026-09-10artificialanalysis.ai ↗ Evidence Axis: AA Intelligence Index · snapshot-2026-09-10 (unversioned) · Published board Exact value: `45.2` points Observed: 2026-09-10 · publication date: not recorded Observation id: `legacy:aa_intelligence_index:muse-spark-1.3::xhigh` Source: https://artificialanalysis.ai/leaderboards/models Protocol: inspect · file `data/raw/artificialanalysis.json`	Compare ↗
19	Claude Opus 5 (Adaptive Reasoning, Medium Effort) Anthropic · medium	45.1 pointsmeasuredobserved 2026-09-10artificialanalysis.ai ↗ Evidence Axis: AA Intelligence Index · snapshot-2026-09-10 (unversioned) · Published board Exact value: `45.1` points Observed: 2026-09-10 · publication date: not recorded Observation id: `legacy:aa_intelligence_index:claude-opus-5::medium` Source: https://artificialanalysis.ai/leaderboards/models Protocol: inspect · file `data/raw/artificialanalysis.json`	Compare ↗
20	GLM-5.3 (max) Z.ai · open weights · max	44.9 pointsmeasuredobserved 2026-09-10artificialanalysis.ai ↗ Evidence Axis: AA Intelligence Index · snapshot-2026-09-10 (unversioned) · Published board Exact value: `44.9` points Observed: 2026-09-10 · publication date: not recorded Observation id: `legacy:aa_intelligence_index:glm-5.3::max` Source: https://artificialanalysis.ai/leaderboards/models Protocol: inspect · file `data/raw/artificialanalysis.json`	Compare ↗
21	Grok 4.6 (high) xAI · high	44.4 pointsmeasuredobserved 2026-09-10artificialanalysis.ai ↗ Evidence Axis: AA Intelligence Index · snapshot-2026-09-10 (unversioned) · Published board Exact value: `44.4` points Observed: 2026-09-10 · publication date: not recorded Observation id: `legacy:aa_intelligence_index:grok-4.6::high` Source: https://artificialanalysis.ai/leaderboards/models Protocol: inspect · file `data/raw/artificialanalysis.json`	Compare ↗
22	Grok 4.6 (xhigh) xAI · xhigh	44.3 pointsmeasuredobserved 2026-09-10artificialanalysis.ai ↗ Evidence Axis: AA Intelligence Index · snapshot-2026-09-10 (unversioned) · Published board Exact value: `44.3` points Observed: 2026-09-10 · publication date: not recorded Observation id: `legacy:aa_intelligence_index:grok-4.6::xhigh` Source: https://artificialanalysis.ai/leaderboards/models Protocol: inspect · file `data/raw/artificialanalysis.json`	Compare ↗
23	GPT-5.6 Sol (xhigh) OpenAI · xhigh	44.1 pointsmeasuredobserved 2026-09-10artificialanalysis.ai ↗ Evidence Axis: AA Intelligence Index · snapshot-2026-09-10 (unversioned) · Published board Exact value: `44.1` points Observed: 2026-09-10 · publication date: not recorded Observation id: `legacy:aa_intelligence_index:gpt-5.6-sol::xhigh` Source: https://artificialanalysis.ai/leaderboards/models Protocol: inspect · file `data/raw/artificialanalysis.json`	Compare ↗
24	Kimi K3 (max) Moonshot AI · open weights · max	43.8 pointsmeasuredobserved 2026-09-10artificialanalysis.ai ↗ Evidence Axis: AA Intelligence Index · snapshot-2026-09-10 (unversioned) · Published board Exact value: `43.8` points Observed: 2026-09-10 · publication date: not recorded Observation id: `legacy:aa_intelligence_index:kimi-k3::max` Source: https://artificialanalysis.ai/leaderboards/models Protocol: inspect · file `data/raw/artificialanalysis.json`	Compare ↗
25	Grok 4.6 (medium) xAI · medium	43 pointsmeasuredobserved 2026-09-10artificialanalysis.ai ↗ Evidence Axis: AA Intelligence Index · snapshot-2026-09-10 (unversioned) · Published board Exact value: `43` points Observed: 2026-09-10 · publication date: not recorded Observation id: `legacy:aa_intelligence_index:grok-4.6::medium` Source: https://artificialanalysis.ai/leaderboards/models Protocol: inspect · file `data/raw/artificialanalysis.json`	Compare ↗
26	GPT-5.6 Sol (high) OpenAI · high	42.5 pointsmeasuredobserved 2026-09-10artificialanalysis.ai ↗ Evidence Axis: AA Intelligence Index · snapshot-2026-09-10 (unversioned) · Published board Exact value: `42.5` points Observed: 2026-09-10 · publication date: not recorded Observation id: `legacy:aa_intelligence_index:gpt-5.6-sol::high` Source: https://artificialanalysis.ai/leaderboards/models Protocol: inspect · file `data/raw/artificialanalysis.json`	Compare ↗
27	GPT-5.6 Terra (max) OpenAI · max	42.3 pointsmeasuredobserved 2026-09-10artificialanalysis.ai ↗ Evidence Axis: AA Intelligence Index · snapshot-2026-09-10 (unversioned) · Published board Exact value: `42.3` points Observed: 2026-09-10 · publication date: not recorded Observation id: `legacy:aa_intelligence_index:gpt-5.6-terra::max` Source: https://artificialanalysis.ai/leaderboards/models Protocol: inspect · file `data/raw/artificialanalysis.json`	Compare ↗
28	Claude Opus 4.8 (Adaptive Reasoning, Max Effort) Anthropic · max	42 pointsmeasuredobserved 2026-09-10artificialanalysis.ai ↗ Evidence Axis: AA Intelligence Index · snapshot-2026-09-10 (unversioned) · Published board Exact value: `42` points Observed: 2026-09-10 · publication date: not recorded Observation id: `legacy:aa_intelligence_index:claude-opus-4.8::max` Source: https://artificialanalysis.ai/leaderboards/models Protocol: inspect · file `data/raw/artificialanalysis.json`	Compare ↗
29	GLM-5.3-Flash Z.ai · open weights · default	41.9 pointsmeasuredobserved 2026-09-10artificialanalysis.ai ↗ Evidence Axis: AA Intelligence Index · snapshot-2026-09-10 (unversioned) · Published board Exact value: `41.9` points Observed: 2026-09-10 · publication date: not recorded Observation id: `legacy:aa_intelligence_index:glm-5.3-flash::default` Source: https://artificialanalysis.ai/leaderboards/models Protocol: inspect · file `data/raw/artificialanalysis.json`	Compare ↗
30	Gemini 3.8 Flash (high) Google · high	41.2 pointsmeasuredobserved 2026-09-10artificialanalysis.ai ↗ Evidence Axis: AA Intelligence Index · snapshot-2026-09-10 (unversioned) · Published board Exact value: `41.2` points Observed: 2026-09-10 · publication date: not recorded Observation id: `legacy:aa_intelligence_index:gemini-3.8-flash::high` Source: https://artificialanalysis.ai/leaderboards/models Protocol: inspect · file `data/raw/artificialanalysis.json`	Compare ↗
31	Claude Opus 4.7 (Adaptive Reasoning, Max Effort) Anthropic · max	40.7 pointsmeasuredobserved 2026-09-10artificialanalysis.ai ↗ Evidence Axis: AA Intelligence Index · snapshot-2026-09-10 (unversioned) · Published board Exact value: `40.7` points Observed: 2026-09-10 · publication date: not recorded Observation id: `legacy:aa_intelligence_index:claude-opus-4.7::max` Source: https://artificialanalysis.ai/leaderboards/models Protocol: inspect · file `data/raw/artificialanalysis.json`	Compare ↗
32	Qwen3.8 Max Alibaba · default	40.3 pointsmeasuredobserved 2026-09-10artificialanalysis.ai ↗ Evidence Axis: AA Intelligence Index · snapshot-2026-09-10 (unversioned) · Published board Exact value: `40.3` points Observed: 2026-09-10 · publication date: not recorded Observation id: `legacy:aa_intelligence_index:qwen3.8-max::default` Source: https://artificialanalysis.ai/leaderboards/models Protocol: inspect · file `data/raw/artificialanalysis.json`	Compare ↗
33	Gemini 3.8 Flash (medium) Google · medium	40 pointsmeasuredobserved 2026-09-10artificialanalysis.ai ↗ Evidence Axis: AA Intelligence Index · snapshot-2026-09-10 (unversioned) · Published board Exact value: `40` points Observed: 2026-09-10 · publication date: not recorded Observation id: `legacy:aa_intelligence_index:gemini-3.8-flash::medium` Source: https://artificialanalysis.ai/leaderboards/models Protocol: inspect · file `data/raw/artificialanalysis.json`	Compare ↗
33	Qwen3.8 2.4T A95B Alibaba · open weights · default	40 pointsmeasuredobserved 2026-09-10artificialanalysis.ai ↗ Evidence Axis: AA Intelligence Index · snapshot-2026-09-10 (unversioned) · Published board Exact value: `40` points Observed: 2026-09-10 · publication date: not recorded Observation id: `legacy:aa_intelligence_index:qwen3.8-2.4t-a95b::default` Source: https://artificialanalysis.ai/leaderboards/models Protocol: inspect · file `data/raw/artificialanalysis.json`	Compare ↗
35	Qwen3.8-Flash-Next Alibaba · open weights · default	39.9 pointsmeasuredobserved 2026-09-10artificialanalysis.ai ↗ Evidence Axis: AA Intelligence Index · snapshot-2026-09-10 (unversioned) · Published board Exact value: `39.9` points Observed: 2026-09-10 · publication date: not recorded Observation id: `legacy:aa_intelligence_index:qwen3.8-flash-next::default` Source: https://artificialanalysis.ai/leaderboards/models Protocol: inspect · file `data/raw/artificialanalysis.json`	Compare ↗
36	Claude Opus 5 (Adaptive Reasoning, Low Effort) Anthropic · low	39.8 pointsmeasuredobserved 2026-09-10artificialanalysis.ai ↗ Evidence Axis: AA Intelligence Index · snapshot-2026-09-10 (unversioned) · Published board Exact value: `39.8` points Observed: 2026-09-10 · publication date: not recorded Observation id: `legacy:aa_intelligence_index:claude-opus-5::low` Source: https://artificialanalysis.ai/leaderboards/models Protocol: inspect · file `data/raw/artificialanalysis.json`	Compare ↗
36	Muse Spark 1.2 (xhigh) Meta · xhigh	39.8 pointsmeasuredobserved 2026-09-10artificialanalysis.ai ↗ Evidence Axis: AA Intelligence Index · snapshot-2026-09-10 (unversioned) · Published board Exact value: `39.8` points Observed: 2026-09-10 · publication date: not recorded Observation id: `legacy:aa_intelligence_index:muse-spark-1.2::xhigh` Source: https://artificialanalysis.ai/leaderboards/models Protocol: inspect · file `data/raw/artificialanalysis.json`	Compare ↗
38	Gemini 3.7 Flash (medium) Google · medium	39.6 pointsmeasuredobserved 2026-09-10artificialanalysis.ai ↗ Evidence Axis: AA Intelligence Index · snapshot-2026-09-10 (unversioned) · Published board Exact value: `39.6` points Observed: 2026-09-10 · publication date: not recorded Observation id: `legacy:aa_intelligence_index:gemini-3.7-flash::medium` Source: https://artificialanalysis.ai/leaderboards/models Protocol: inspect · file `data/raw/artificialanalysis.json`	Compare ↗
39	GPT-5.6 Sol (medium) OpenAI · medium	39.5 pointsmeasuredobserved 2026-09-10artificialanalysis.ai ↗ Evidence Axis: AA Intelligence Index · snapshot-2026-09-10 (unversioned) · Published board Exact value: `39.5` points Observed: 2026-09-10 · publication date: not recorded Observation id: `legacy:aa_intelligence_index:gpt-5.6-sol::medium` Source: https://artificialanalysis.ai/leaderboards/models Protocol: inspect · file `data/raw/artificialanalysis.json`	Compare ↗
40	Gemini 3.7 Flash (high) Google · high	39.4 pointsmeasuredobserved 2026-09-10artificialanalysis.ai ↗ Evidence Axis: AA Intelligence Index · snapshot-2026-09-10 (unversioned) · Published board Exact value: `39.4` points Observed: 2026-09-10 · publication date: not recorded Observation id: `legacy:aa_intelligence_index:gemini-3.7-flash::high` Source: https://artificialanalysis.ai/leaderboards/models Protocol: inspect · file `data/raw/artificialanalysis.json`	Compare ↗
41	Grok 4.5 (high) xAI · high	39.1 pointsmeasuredobserved 2026-09-10artificialanalysis.ai ↗ Evidence Axis: AA Intelligence Index · snapshot-2026-09-10 (unversioned) · Published board Exact value: `39.1` points Observed: 2026-09-10 · publication date: not recorded Observation id: `legacy:aa_intelligence_index:grok-4.5::high` Source: https://artificialanalysis.ai/leaderboards/models Protocol: inspect · file `data/raw/artificialanalysis.json`	Compare ↗
42	GPT-5.4 (xhigh) OpenAI · xhigh	39 pointsmeasuredobserved 2026-09-10artificialanalysis.ai ↗ Evidence Axis: AA Intelligence Index · snapshot-2026-09-10 (unversioned) · Published board Exact value: `39` points Observed: 2026-09-10 · publication date: not recorded Observation id: `legacy:aa_intelligence_index:gpt-5.4::xhigh` Source: https://artificialanalysis.ai/leaderboards/models Protocol: inspect · file `data/raw/artificialanalysis.json`	Compare ↗
43	GLM-5.2 (max) Z.ai · open weights · max	38.6 pointsmeasuredobserved 2026-09-10artificialanalysis.ai ↗ Evidence Axis: AA Intelligence Index · snapshot-2026-09-10 (unversioned) · Published board Exact value: `38.6` points Observed: 2026-09-10 · publication date: not recorded Observation id: `legacy:aa_intelligence_index:glm-5.2::max` Source: https://artificialanalysis.ai/leaderboards/models Protocol: inspect · file `data/raw/artificialanalysis.json`	Compare ↗
43	GPT-5.5 (xhigh) OpenAI · xhigh	38.6 pointsmeasuredobserved 2026-09-10artificialanalysis.ai ↗ Evidence Axis: AA Intelligence Index · snapshot-2026-09-10 (unversioned) · Published board Exact value: `38.6` points Observed: 2026-09-10 · publication date: not recorded Observation id: `legacy:aa_intelligence_index:gpt-5.5::xhigh` Source: https://artificialanalysis.ai/leaderboards/models Protocol: inspect · file `data/raw/artificialanalysis.json`	Compare ↗
45	Claude Sonnet 5 (Adaptive Reasoning, Max Effort) Anthropic · max	38.4 pointsmeasuredobserved 2026-09-10artificialanalysis.ai ↗ Evidence Axis: AA Intelligence Index · snapshot-2026-09-10 (unversioned) · Published board Exact value: `38.4` points Observed: 2026-09-10 · publication date: not recorded Observation id: `legacy:aa_intelligence_index:claude-sonnet-5::max` Source: https://artificialanalysis.ai/leaderboards/models Protocol: inspect · file `data/raw/artificialanalysis.json`	Compare ↗
46	GPT-5.6 Terra (xhigh) OpenAI · xhigh	38.2 pointsmeasuredobserved 2026-09-10artificialanalysis.ai ↗ Evidence Axis: AA Intelligence Index · snapshot-2026-09-10 (unversioned) · Published board Exact value: `38.2` points Observed: 2026-09-10 · publication date: not recorded Observation id: `legacy:aa_intelligence_index:gpt-5.6-terra::xhigh` Source: https://artificialanalysis.ai/leaderboards/models Protocol: inspect · file `data/raw/artificialanalysis.json`	Compare ↗
47	GPT-5.6 Luna (max) OpenAI · max	37.5 pointsmeasuredobserved 2026-09-10artificialanalysis.ai ↗ Evidence Axis: AA Intelligence Index · snapshot-2026-09-10 (unversioned) · Published board Exact value: `37.5` points Observed: 2026-09-10 · publication date: not recorded Observation id: `legacy:aa_intelligence_index:gpt-5.6-luna::max` Source: https://artificialanalysis.ai/leaderboards/models Protocol: inspect · file `data/raw/artificialanalysis.json`	Compare ↗
48	GPT-5.5 (high) OpenAI · high	37.3 pointsmeasuredobserved 2026-09-10artificialanalysis.ai ↗ Evidence Axis: AA Intelligence Index · snapshot-2026-09-10 (unversioned) · Published board Exact value: `37.3` points Observed: 2026-09-10 · publication date: not recorded Observation id: `legacy:aa_intelligence_index:gpt-5.5::high` Source: https://artificialanalysis.ai/leaderboards/models Protocol: inspect · file `data/raw/artificialanalysis.json`	Compare ↗
49	Gemini 3.7 Flash (low) Google · low	36.9 pointsmeasuredobserved 2026-09-10artificialanalysis.ai ↗ Evidence Axis: AA Intelligence Index · snapshot-2026-09-10 (unversioned) · Published board Exact value: `36.9` points Observed: 2026-09-10 · publication date: not recorded Observation id: `legacy:aa_intelligence_index:gemini-3.7-flash::low` Source: https://artificialanalysis.ai/leaderboards/models Protocol: inspect · file `data/raw/artificialanalysis.json`	Compare ↗
50	DeepSeek V4 Pro 0813 (Reasoning, Max Effort) DeepSeek · open weights · max	36.3 pointsmeasuredobserved 2026-09-10artificialanalysis.ai ↗ Evidence Axis: AA Intelligence Index · snapshot-2026-09-10 (unversioned) · Published board Exact value: `36.3` points Observed: 2026-09-10 · publication date: not recorded Observation id: `legacy:aa_intelligence_index:deepseek-v4-pro-0813::max` Source: https://artificialanalysis.ai/leaderboards/models Protocol: inspect · file `data/raw/artificialanalysis.json`	Compare ↗

Ranks stay within this version and evaluation group. Model prompts, evaluation dates and reasoning settings may differ. Source identities can describe whole agent systems. Inspect evidence before treating results as directly equivalent. Statistical significance is not inferred from rank.

Composite retains Coding Agent v1.4 · source 2026-09-09

The five Composite inputs remain unchanged. Its Coding Agent input is the median across complete harness results in the retained v1.4 snapshot from 2026-09-09. Artificial Analysis now publishes v1.5, with different components. Explore current v1.5 separately. Dated snapshot labels on other indices identify unversioned source captures, not a verified semantic version.