Skip to content

Benchmark

Artificial Analysis Briefcase

Scores are Elo ratings. Higher ratings are better.

Models and systems measured
19
Scores recorded
30
Sources
2
IQ weight
1.34

Score distribution

1,526.1
Mean
1,574
Median
258.4
Std dev
1,336
10th percentile
1,780
90th percentile
614
Min
6141,824
The band covers the middle 80% of scores. The line marks the median.

This benchmark has a weight of 1.34 for chat models, measured over 8 models. The weight is how much it counts toward llminfo IQ. We measure it from the data, not from the raw scores.

Leaderboard

#Model or systemScore
1Claude Sonnet 5.5adaptive thinking · max effort · default sampling · 5-trial mean · context ≤1M unless noted · v1.1 · Artificial Analysis · Haiku 5.5 card snapshot · max· Reported by Anthropic· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.1,824
2Claude Opus 5.5Opus 5.5 System Card capability section · v1.1 · Artificial Analysis · max effort· Reported by Anthropic· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.1,822
3Claude Sonnet 5.5v1.1 · Artificial Analysis · max· Reported by Anthropic· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.1,811
4Claude Opus 5.5Opus 5.5 System Card capability section · v1.1 · Artificial Analysis · xhigh effort· Reported by Anthropic· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.1,780
5Claude Sonnet 5.5adaptive thinking · max effort · default sampling · 5-trial mean · context ≤1M unless noted · v1.1 · Artificial Analysis · xhigh· Reported by Anthropic· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.1,746
6Claude Opus 5Opus 5 System Card capability summary · adaptive thinking · max effort unless noted· Reported by Anthropic· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.1,720
7Claude Opus 5.5Opus 5.5 System Card capability section · v1.1 · Artificial Analysis · high effort· Reported by Anthropic· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.1,705
8Claude Fable 5.1System Card capability summary · adaptive thinking · max effort · default sampling · 5 trials · context ≤1M unless noted· Reported by Anthropic· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.1,694
9Claude Opus 5System Card capability summary · adaptive thinking · max effort · default sampling · 5 trials · context ≤1M unless noted· Reported by Anthropic· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.1,685
10Claude Fable 5.1Opus 5.5 System Card capability section · v1.1 · Artificial Analysis · max effort· Reported by Anthropic· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.1,678
11Claude Opus 5Opus 5.5 System Card capability section · v1.1 · Artificial Analysis · max effort· Reported by Anthropic· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.1,673
12Grok 4.7release-page snapshot· Reported by SpaceXAI· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.1,657
13Claude Haiku 5.5v1.1 · Artificial Analysis · max· Reported by Anthropic· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.1,578
14Grok 4.6· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim1,577
15Claude Fable 5Opus 5 System Card capability summary · adaptive thinking · max effort unless noted· Reported by Anthropic· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.1,574
16Claude Fable 5System Card capability summary · adaptive thinking · max effort · default sampling · 5 trials · context ≤1M unless noted· Reported by Anthropic· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.1,572
17Kimi K3· LLM Stats data1,548
18Kimi K3snapshot 2026-07-23· Reported by Moonshot AI· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.1,548
19Inkling 2 Small· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim1,527
20GPT-5.6 SolOpus 5 System Card capability summary · adaptive thinking · max effort unless noted· Reported by Anthropic· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.1,505
21GPT-5.6 SolSystem Card capability summary · adaptive thinking · max effort · default sampling · 5 trials · context ≤1M unless noted· Reported by Anthropic· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.1,502
22Claude Sonnet 5Elo · Artificial Analysis independent evaluation· Reported by Anthropic· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.1,393
23Mistral Large 4· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim1,393
24Claude Haiku 5.5adaptive thinking · max effort · default sampling · 5-trial mean · context ≤1M unless noted · v1.1 · Artificial Analysis · medium· Reported by Anthropic· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.1,372
25Claude Opus 4.8Opus 5 System Card capability summary · adaptive thinking · max effort unless noted· Reported by Anthropic· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.1,346
26GPT-6 Lunaadaptive thinking · max effort · default sampling · 5-trial mean · context ≤1M unless noted · v1.1 · Artificial Analysis · Haiku 5.5 card snapshot · max· Reported by Anthropic· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.1,336
27Gemini 3.8 Flashadaptive thinking · max effort · default sampling · 5-trial mean · context ≤1M unless noted · v1.1 · Artificial Analysis · Haiku 5.5 card snapshot · highest thinking· Reported by Anthropic· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.1,203
28Inkling Small· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim917
29Claude Haiku 4.5adaptive thinking · max effort · default sampling · 5-trial mean · context ≤1M unless noted · v1.1 · Artificial Analysis · published setting· Reported by Anthropic· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.614

Not ranked

The source marks these results as descriptive, not capability scores, so we list them by name instead of ranking them.

Model or systemScore
GPT-6 Sol (Sonnet 5.5 table label)source table label · max effort· Reported by Anthropic· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.1,483

Data credits: BenchBoard by SYuan03 (github.com/SYuan03/benchboard). MIT covers its code and data organization; no results licence is inferred. Numeric relays retain the original stating-source links. · Data by LLM Stats