Skip to main content
AI Choice Engine
← Benchmark sources

artificialanalysis.ai · benchmark source

Artificial Analysis

Independent evaluations across composite intelligence, hard reasoning, science reasoning, and agentic terminal tasks. Multiple boards live under one publisher.

Open artificialanalysis.ai5 boards on this publisher

General capability

Intelligence Index

A composite index aggregating several evaluations into a comparative model-quality signal.

Why it matters

It is useful as one broad reference point when a buyer needs an initial shortlist.

Use it when

You need an orienting signal, then plan to validate candidates on your own workload.

The limitation

Composite scores hide task-specific variance and the trade-off between quality and cost. Version changes rebaseline the scale — v4.3.2 scores are not comparable with v4.1.1.

Leaderboard snapshot

Intelligence Index results in a readable view

Use the tabs for the primary score and other published fields from this board. The source link remains authoritative for the live table.

Intelligence Index v4.3.2 is rebaselined and not comparable with v4.1.1. Only unique exact public-table matches are retained; conflicting source scores are excluded. AA estimates remain labelled per row.
115 exact matching primary-source observations are ranked in this retained subset. 148 unsupported observations are excluded from ranks, charts and leader claims. Subset ranks are not positions on the complete publisher leaderboard.

Leader among verified rows

Claude Opus 5.5 (max with fallback)

58

Rows shown

102

Index score

Source / snapshot date

2026-10-01

25 days old · not a live API feed

Score profile

Index score by published row

Higher is better in this view

Claude Opus 5.5 (max with fallback)
58
Claude Opus 5.5 (xhigh with fallback)
56
Claude Sonnet 5.5 (max with fallback)
56
Claude Opus 5.5 (high with fallback)
54
Claude Fable 5.1 (max with fallback)
53
Claude Fable 5.1 (xhigh with fallback)
53
Gemini 4 Argon (high)
53
GPT-6 Astra (max)
53
Claude Sonnet 5.5 (xhigh with fallback)
52
GPT-6 Astra (xhigh)
52
GPT-6.1 Sol (max)
52
Claude Fable 5.1 (high with fallback)
51
Claude Opus 5.5 (medium with fallback)
51
GPT-6 Astra (high)
51
GPT-6.1 Sol (xhigh)
51
GPT-6 Astra (medium)
50
GPT-6.1 Sol (high)
50
Claude Fable 5.1 (medium with fallback)
49
GPT-6 Sol (max)
48
GPT-6.1 Sol (medium)
48
Muse Spark 1.3 (max)
48
Claude Fable 5.1 (low with fallback)
47
Claude Sonnet 5.5 (high with fallback)
47
GPT-6 Astra (low)
46
Grok 4.7 (high)
46
Grok 4.7 (xhigh)
46
MiMo-V2.6-Pro
46
GLM-5.3 (max)
45
Muse Spark 1.3 (xhigh)
45
Qwen3.8 Max (0902)
45
GPT-6 Sol (xhigh)
44
Kimi K3 (max)
44
Step 5 Preview
44
Claude Opus 5.5 (low with fallback)
42
GLM-5.3-Flash
42
GPT-5.6 Terra (max)
42
GPT-6.1 Sol (low)
42
Claude Sonnet 5.5 (medium with fallback)
41
Gemini 3.8 Flash (high)
41
Gemini 3.8 Flash (medium)
40
GPT-6 Sol (medium)
40
Qwen3.8 2.4T A95B
40
Qwen3.8-Flash-Next
40
DeepSeek V4.1 Flash (max)
39
GPT-5.6 Terra (xhigh)
38
MiMo-V2.6-Flash
38
DeepSeek V4 Pro 0813 (max)
36
DeepSeek V4 Flash Vision (max)
35
GLM-5.3 (low)
34
GPT-5.6 Terra (high)
34
GPT-6 Sol (low)
34
JT-4.1 Flash 236B A21B
34
Qwen3.8 27B (xhigh)
34
Gemini 3.8 Flash (low)
33
K2 Horizon 375B A23B
31
Gemini 3.1 Pro Preview
30
GPT-5.6 Terra (medium)
30
Kimi K3 (low)
30
MiniMax-M3
29
Qwen3.8 27B (medium)
28
GPT-5.6 Terra (low)
27
Quasar 438B (max)
27
Apodex 1.1
26
GPT-5.5 Instant (June 2026)
26
Inkling Small
26
Kimi K2.7 Code
26
Qwen3.8 27B (low)
26
DeepSeek V4.1 Flash (non-reasoning)
25
Hy3
25
Inkling (xhigh)
25
K2 Horizon MoVA 36B A4B
25
Ling-3.0-flash-VL
25
Qwen3.7 Plus
25
Solar Mini 4
24
Ling-3.0-flash-Fin
23
Nemotron 3 Ultra
23
Gemini 3.5 Flash-Lite
22
GPT-5.6 Terra (non-reasoning)
21
K2 Horizon 7B
21
DeepSeek V4 Pro 0813 (non-reasoning)
20
Ling 3.0 Flash
20
Qwen3.8 27B (non-reasoning)
20
LongCat 2.0
19
GPT-6 Luna (non-reasoning)
18
Qwen3.5 397B A17B
18
Qwen3.6 35B A3B
18
Claude 4.5 Haiku
17
Muse Glimmer (high)
17
Ring-2.6-1T
17
K2 Horizon 3.7B
16
Qwen3.5 122B A10B
16
Gemma 4 31B
15
Mistral Medium 3.5
14
Command A+
13
Nemotron 3 Super
13
Nemotron 3.5 Lightning
13
gpt-oss-120b (high)
12
Mercury 2.5
12
MiniCPM5-2B
12
Granite 4.2 8B
11
Mistral Small 4
11
Mistral Large 3
9
Not yet verified (148)

No exact matching primary-source observation was retained for these entries. Scores are withheld here; these entries do not affect ranks, leaders, charts or search indexing.

  • GPT-6 Sol (high)

    No exact matching primary-source row was retained; excluded from rankings.

  • GPT-6 Luna (max)

    No exact matching primary-source row was retained; excluded from rankings.

  • GPT-6 Luna (xhigh)

    No exact matching primary-source row was retained; excluded from rankings.

  • Motif 3

    Artificial Analysis estimate, as marked by the publisher. No exact matching primary-source row was retained; excluded from rankings.

  • GPT-5.3 Codex (xhigh)

    Artificial Analysis estimate, as marked by the publisher. No exact matching primary-source row was retained; excluded from rankings.

  • GPT-6 Luna (high)

    No exact matching primary-source row was retained; excluded from rankings.

  • GPT-6 Luna (medium)

    No exact matching primary-source row was retained; excluded from rankings.

  • GPT-6 Sol (non-reasoning)

    No exact matching primary-source row was retained; excluded from rankings.

  • Nex-N2-Pro

    Artificial Analysis estimate, as marked by the publisher. No exact matching primary-source row was retained; excluded from rankings.

  • Solar Pro 4

    Artificial Analysis estimate, as marked by the publisher. No exact matching primary-source row was retained; excluded from rankings.

  • JT-4.1 Flash 236B A21B (non-reasoning)

    Artificial Analysis estimate, as marked by the publisher. No exact matching primary-source row was retained; excluded from rankings.

  • Solar Open2 250B

    Artificial Analysis estimate, as marked by the publisher. No exact matching primary-source row was retained; excluded from rankings.

  • A.X-K2

    Artificial Analysis estimate, as marked by the publisher. No exact matching primary-source row was retained; excluded from rankings.

  • G9v3-39A5B

    Artificial Analysis estimate, as marked by the publisher. No exact matching primary-source row was retained; excluded from rankings.

  • GPT-6 Luna (low)

    No exact matching primary-source row was retained; excluded from rankings.

  • KAT-Coder-Pro V2

    Artificial Analysis estimate, as marked by the publisher. No exact matching primary-source row was retained; excluded from rankings.

  • Qwen3.5 397B A17B (non-reasoning)

    Artificial Analysis estimate, as marked by the publisher. No exact matching primary-source row was retained; excluded from rankings.

  • K-EXAONE 2.0

    Artificial Analysis estimate, as marked by the publisher. No exact matching primary-source row was retained; excluded from rankings.

  • o3

    Artificial Analysis estimate, as marked by the publisher. No exact matching primary-source row was retained; excluded from rankings.

  • Qwen3.5 Omni Plus

    Artificial Analysis estimate, as marked by the publisher. No exact matching primary-source row was retained; excluded from rankings.

  • JT-35B-Flash

    Artificial Analysis estimate, as marked by the publisher. No exact matching primary-source row was retained; excluded from rankings.

  • Qwen3.5 122B A10B (non-reasoning)

    Artificial Analysis estimate, as marked by the publisher. No exact matching primary-source row was retained; excluded from rankings.

  • Doubao Seed Code

    Artificial Analysis estimate, as marked by the publisher. No exact matching primary-source row was retained; excluded from rankings.

  • Gemma 4 26B A4B

    Artificial Analysis estimate, as marked by the publisher. No exact matching primary-source row was retained; excluded from rankings.

  • Claude 4.5 Haiku (non-reasoning)

    Artificial Analysis estimate, as marked by the publisher. No exact matching primary-source row was retained; excluded from rankings.

  • Granite 4.2 30B

    Artificial Analysis estimate, as marked by the publisher. No exact matching primary-source row was retained; excluded from rankings.

  • Qwen3.6 35B A3B (non-reasoning)

    Artificial Analysis estimate, as marked by the publisher. No exact matching primary-source row was retained; excluded from rankings.

  • ERNIE 5.0 Thinking Preview

    Artificial Analysis estimate, as marked by the publisher. No exact matching primary-source row was retained; excluded from rankings.

  • Gemma 4 12B

    Artificial Analysis estimate, as marked by the publisher. No exact matching primary-source row was retained; excluded from rankings.

  • Gemma 4 31B (non-reasoning)

    Artificial Analysis estimate, as marked by the publisher. No exact matching primary-source row was retained; excluded from rankings.

  • Nova 2.0 Omni (medium)

    Artificial Analysis estimate, as marked by the publisher. No exact matching primary-source row was retained; excluded from rankings.

  • Nova 2.0 Pro Preview (medium)

    Artificial Analysis estimate, as marked by the publisher. No exact matching primary-source row was retained; excluded from rankings.

  • Apriel-v1.6-15B-Thinker

    Artificial Analysis estimate, as marked by the publisher. No exact matching primary-source row was retained; excluded from rankings.

  • EXAONE 4.5 33B

    Artificial Analysis estimate, as marked by the publisher. No exact matching primary-source row was retained; excluded from rankings.

  • Gemma 4 26B A4B (non-reasoning)

    Artificial Analysis estimate, as marked by the publisher. No exact matching primary-source row was retained; excluded from rankings.

  • Nova 2.0 Lite (high)

    Artificial Analysis estimate, as marked by the publisher. No exact matching primary-source row was retained; excluded from rankings.

  • Nova 2.0 Pro Preview (low)

    Artificial Analysis estimate, as marked by the publisher. No exact matching primary-source row was retained; excluded from rankings.

  • Qwen3.5 4B

    Artificial Analysis estimate, as marked by the publisher. No exact matching primary-source row was retained; excluded from rankings.

  • Qwen3.5 9B (non-reasoning)

    Artificial Analysis estimate, as marked by the publisher. No exact matching primary-source row was retained; excluded from rankings.

  • HyperNova 60B 2605 (high)

    Artificial Analysis estimate, as marked by the publisher. No exact matching primary-source row was retained; excluded from rankings.

  • JT-MINI

    Artificial Analysis estimate, as marked by the publisher. No exact matching primary-source row was retained; excluded from rankings.

  • Magistral Medium 1.2

    Artificial Analysis estimate, as marked by the publisher. No exact matching primary-source row was retained; excluded from rankings.

  • Nemotron Cascade 2 30B A3B

    Artificial Analysis estimate, as marked by the publisher. No exact matching primary-source row was retained; excluded from rankings.

  • Nova 2.0 Lite (low)

    Artificial Analysis estimate, as marked by the publisher. No exact matching primary-source row was retained; excluded from rankings.

  • Nova 2.0 Lite (medium)

    Artificial Analysis estimate, as marked by the publisher. No exact matching primary-source row was retained; excluded from rankings.

  • Qwen3.5 Omni Flash

    Artificial Analysis estimate, as marked by the publisher. No exact matching primary-source row was retained; excluded from rankings.

  • G9v3-3B

    Artificial Analysis estimate, as marked by the publisher. No exact matching primary-source row was retained; excluded from rankings.

  • HyperCLOVA X SEED Think (32B)

    Artificial Analysis estimate, as marked by the publisher. No exact matching primary-source row was retained; excluded from rankings.

  • INTELLECT-3

    Artificial Analysis estimate, as marked by the publisher. No exact matching primary-source row was retained; excluded from rankings.

  • K2 Think V2

    Artificial Analysis estimate, as marked by the publisher. No exact matching primary-source row was retained; excluded from rankings.

  • LongCat Flash Lite

    Artificial Analysis estimate, as marked by the publisher. No exact matching primary-source row was retained; excluded from rankings.

  • Mi:dm K 2.5 Pro

    Artificial Analysis estimate, as marked by the publisher. No exact matching primary-source row was retained; excluded from rankings.

  • Nova 2.0 Omni (low)

    Artificial Analysis estimate, as marked by the publisher. No exact matching primary-source row was retained; excluded from rankings.

  • Qwen3 Next 80B A3B

    Artificial Analysis estimate, as marked by the publisher. No exact matching primary-source row was retained; excluded from rankings.

  • Qwen3.5 4B (non-reasoning)

    Artificial Analysis estimate, as marked by the publisher. No exact matching primary-source row was retained; excluded from rankings.

  • DiffusionGemma 26B A4B

    Artificial Analysis estimate, as marked by the publisher. No exact matching primary-source row was retained; excluded from rankings.

  • gpt-oss-120b (low)

    Artificial Analysis estimate, as marked by the publisher. No exact matching primary-source row was retained; excluded from rankings.

  • gpt-oss-20b (low)

    Artificial Analysis estimate, as marked by the publisher. No exact matching primary-source row was retained; excluded from rankings.

  • K2-V2 (high)

    Artificial Analysis estimate, as marked by the publisher. No exact matching primary-source row was retained; excluded from rankings.

  • Llama 4 Maverick

    Artificial Analysis estimate, as marked by the publisher. No exact matching primary-source row was retained; excluded from rankings.

  • Nemotron 3 Nano Omni 30B A3B

    Artificial Analysis estimate, as marked by the publisher. No exact matching primary-source row was retained; excluded from rankings.

  • Nova 2.0 Pro Preview (non-reasoning)

    Artificial Analysis estimate, as marked by the publisher. No exact matching primary-source row was retained; excluded from rankings.

  • Qwen3 Next 80B A3B

    Artificial Analysis estimate, as marked by the publisher. No exact matching primary-source row was retained; excluded from rankings.

  • Solar Open 100B

    Artificial Analysis estimate, as marked by the publisher. No exact matching primary-source row was retained; excluded from rankings.

  • Gemma 4 12B (non-reasoning)

    Artificial Analysis estimate, as marked by the publisher. No exact matching primary-source row was retained; excluded from rankings.

  • Gemma 4 E4B

    Artificial Analysis estimate, as marked by the publisher. No exact matching primary-source row was retained; excluded from rankings.

  • K2-V2 (medium)

    Artificial Analysis estimate, as marked by the publisher. No exact matching primary-source row was retained; excluded from rankings.

  • Magistral Small 1.2

    Artificial Analysis estimate, as marked by the publisher. No exact matching primary-source row was retained; excluded from rankings.

  • MiniCPM5-1B

    Artificial Analysis estimate, as marked by the publisher. No exact matching primary-source row was retained; excluded from rankings.

  • MiniCPM5-1B (non-reasoning)

    Artificial Analysis estimate, as marked by the publisher. No exact matching primary-source row was retained; excluded from rankings.

  • Mistral Small 4 (non-reasoning)

    Artificial Analysis estimate, as marked by the publisher. No exact matching primary-source row was retained; excluded from rankings.

  • Motif-2-12.7B

    Artificial Analysis estimate, as marked by the publisher. No exact matching primary-source row was retained; excluded from rankings.

  • Nova 2.0 Lite (non-reasoning)

    Artificial Analysis estimate, as marked by the publisher. No exact matching primary-source row was retained; excluded from rankings.

  • Nova Premier

    Artificial Analysis estimate, as marked by the publisher. No exact matching primary-source row was retained; excluded from rankings.

  • Sarvam 105B (high)

    Artificial Analysis estimate, as marked by the publisher. No exact matching primary-source row was retained; excluded from rankings.

  • Tri-21B-Think

    Artificial Analysis estimate, as marked by the publisher. No exact matching primary-source row was retained; excluded from rankings.

  • ERNIE 4.5 300B A47B

    Artificial Analysis estimate, as marked by the publisher. No exact matching primary-source row was retained; excluded from rankings.

  • EXAONE 4.0 32B

    Artificial Analysis estimate, as marked by the publisher. No exact matching primary-source row was retained; excluded from rankings.

  • Falcon-H1R-7B

    Artificial Analysis estimate, as marked by the publisher. No exact matching primary-source row was retained; excluded from rankings.

  • Gemma 4 E2B

    Artificial Analysis estimate, as marked by the publisher. No exact matching primary-source row was retained; excluded from rankings.

  • Hermes 4 70B

    Artificial Analysis estimate, as marked by the publisher. No exact matching primary-source row was retained; excluded from rankings.

  • LFM2.5-2.6B

    Artificial Analysis estimate, as marked by the publisher. No exact matching primary-source row was retained; excluded from rankings.

  • Llama 3.3 70B

    Artificial Analysis estimate, as marked by the publisher. No exact matching primary-source row was retained; excluded from rankings.

  • Llama 4 Scout

    Artificial Analysis estimate, as marked by the publisher. No exact matching primary-source row was retained; excluded from rankings.

  • Llama Nemotron Ultra

    Artificial Analysis estimate, as marked by the publisher. No exact matching primary-source row was retained; excluded from rankings.

  • Nanbeige4.1-3B

    Artificial Analysis estimate, as marked by the publisher. No exact matching primary-source row was retained; excluded from rankings.

  • Nova 2.0 Omni (non-reasoning)

    Artificial Analysis estimate, as marked by the publisher. No exact matching primary-source row was retained; excluded from rankings.

  • Qwen3 Omni 30B A3B

    Artificial Analysis estimate, as marked by the publisher. No exact matching primary-source row was retained; excluded from rankings.

  • Step3 VL 10B

    Artificial Analysis estimate, as marked by the publisher. No exact matching primary-source row was retained; excluded from rankings.

  • Command A

    Artificial Analysis estimate, as marked by the publisher. No exact matching primary-source row was retained; excluded from rankings.

  • Gemma 4 E4B (non-reasoning)

    Artificial Analysis estimate, as marked by the publisher. No exact matching primary-source row was retained; excluded from rankings.

  • Hermes 4 405B

    Artificial Analysis estimate, as marked by the publisher. No exact matching primary-source row was retained; excluded from rankings.

  • Hermes 4 405B (non-reasoning)

    Artificial Analysis estimate, as marked by the publisher. No exact matching primary-source row was retained; excluded from rankings.

  • Hermes 4 70B (non-reasoning)

    Artificial Analysis estimate, as marked by the publisher. No exact matching primary-source row was retained; excluded from rankings.

  • K2-V2 (low)

    Artificial Analysis estimate, as marked by the publisher. No exact matching primary-source row was retained; excluded from rankings.

  • Kimi Linear 48B A3B Instruct

    Artificial Analysis estimate, as marked by the publisher. No exact matching primary-source row was retained; excluded from rankings.

  • LFM2.5-8B-A1B

    Artificial Analysis estimate, as marked by the publisher. No exact matching primary-source row was retained; excluded from rankings.

  • Llama 3.1 405B

    Artificial Analysis estimate, as marked by the publisher. No exact matching primary-source row was retained; excluded from rankings.

  • Nemotron 3 Nano (non-reasoning)

    Artificial Analysis estimate, as marked by the publisher. No exact matching primary-source row was retained; excluded from rankings.

  • Nemotron 3 Nano 4B

    Artificial Analysis estimate, as marked by the publisher. No exact matching primary-source row was retained; excluded from rankings.

Showing 100 entries. Open the publisher source for the current complete table.

Intelligence Index v4.3.2 is rebaselined and not comparable with v4.1.1. Only unique exact public-table matches are retained; conflicting source scores are excluded. AA estimates remain labelled per row. 115 exact matching primary-source observations are ranked in this retained subset. 148 unsupported observations are excluded from ranks, charts and leader claims. Subset ranks are not positions on the complete publisher leaderboard. Exact retained receipts captured 2026-10-01; source/run dates are shown per verified row when known.

Open Artificial Analysis Intelligence Index

Showing the top 100 of 115 entries plus every model with a page in our catalog. 13 lower-ranked entries are on the full source table.

Evidence in this catalog

Where Intelligence Index fits

Closest verified examples we currently carry. A missing score is not a zero.

Model examples

Harness examples

No directly comparable harness score is verified here yet.

Open this board on Artificial Analysis Intelligence Index