Artificial Analysis Briefcase (AA Briefcase)
We show this table for reference; we do not rank on it.
An independently evaluated professional-work benchmark reported as Elo.
Benchmark score on AA Briefcase — November 13, 2026
We compile the AA Briefcase rows from provider self-reports and secondary reports. Claude Sonnet 5.5 leads the table at 1811, followed by Claude Opus 5.5 (1801) and Claude Fable 5.2 (1785). We do not use these results to rank models overall.
Claude Sonnet 5.5
Anthropic
Reported by AnthropicClaude Opus 5.5
Anthropic
Carried from an upstream pageClaude Fable 5.2
Anthropic
Reported by Anthropic23 modelsAgenticCurrentDisplay onlyUpdated November 13, 2026
Benchmark score table (23 models)
None of the 23 results on this page has a documented run by the benchmark or an independent evaluator: 5 reported by the labs that built the models and 18 carried from an upstream page, where who ran them is not recorded.
| Rank | Model / configuration | Score | Who ran it | Parameters (B) | Open / closed |
|---|---|---|---|---|---|
| 1 | Claude Sonnet 5.5AnthropicReported by Anthropic | 1811 | Reported by Anthropic | Not reported | Closed |
| 2 | Claude Opus 5.5AnthropicCarried from an upstream page | 1801 | Carried from an upstream page | Not reported | Closed |
| 3 | Claude Fable 5.2AnthropicReported by Anthropic | 1785 | Reported by Anthropic | Not reported | Closed |
| 4 | Muse Wonder FullMetaCarried from an upstream page | 1729 | Carried from an upstream page | Not reported | Closed |
| 5 | Claude Opus 5AnthropicReported by Anthropic | 1720 | Reported by Anthropic | Not reported | Closed |
| 6 | Kimi K3.1 ProMoonshot AICarried from an upstream page | 1674 | Carried from an upstream page | Not reported | Open |
| 7 | Claude Fable 5.1AnthropicCarried from an upstream page | 1669 | Carried from an upstream page | Not reported | Closed |
| 8 | Muse WonderMetaCarried from an upstream page | 1659 | Carried from an upstream page | Not reported | Closed |
| 9 | Grok 4.7xAIReported by xAI | 1657 | Reported by xAI | Not reported | Closed |
| 10 | Gemini 4 CarbonGoogleCarried from an upstream page | 1611 | Carried from an upstream page | Not reported | Closed |
| 11 | Claude Haiku 5.5AnthropicReported by Anthropic | 1578 | Reported by Anthropic | Not reported | Closed |
| 12 | GPT-6 AstraOpenAICarried from an upstream page | 1569 | Carried from an upstream page | Not reported | Closed |
| 13 | Gemini 4 HydrogenGoogleCarried from an upstream page | 1567 | Carried from an upstream page | Not reported | Closed |
| 14 | GPT-6.1 SolOpenAICarried from an upstream page | 1555 | Carried from an upstream page | Not reported | Closed |
| 15 | Inkling 2 SmallThinking Machines LabCarried from an upstream page | 1527 | Carried from an upstream page | Not reported | Closed |
| 16 | Muse EmberMetaCarried from an upstream page | 1521 | Carried from an upstream page | Not reported | Closed |
| 17 | MiMo-V2.6-ProXiaomiCarried from an upstream page | 1512 | Carried from an upstream page | Not reported | Open |
| 18 | GLM-5.3Z.AICarried from an upstream page | 1503 | Carried from an upstream page | Not reported | Open |
| 19 | GPT-6.1 TerraOpenAICarried from an upstream page | 1403 | Carried from an upstream page | Not reported | Closed |
| 20 | Qwen3.8-27BAlibabaCarried from an upstream page | 1391 | Carried from an upstream page | Not reported | Open |
| 21 | Mistral Large 4MistralCarried from an upstream page | 1348 | Carried from an upstream page | Not reported | Open |
| 22 | GPT-6 LunaOpenAICarried from an upstream page | 1337 | Carried from an upstream page | Not reported | Closed |
| 23 | Beam 1.0 501B-A23BReflection AICarried from an upstream page | 976 | Carried from an upstream page | Not reported | Open |
- Lab-reported
- Benchmark-run
- Independent
Among the reported AA Briefcase rows, Claude Sonnet 5.5 is first at 1811. The third row is 26 score units behind. The broader top-10 range is 200 score units, so the table still separates the published systems.
23 models have been evaluated on AA Briefcase. The benchmark falls in the Agentic category. AA Briefcase is currently displayed for reference but excluded from the scoring formula, so it does not directly affect overall rankings.
About AA Briefcase
Year
2026
Tasks
Professional knowledge-work tasks
Format
Elo
Difficulty
Professional work
Stored as a display-only Elo rating.
Freshness and provenance
Version
AA Briefcase 2026
Refresh cadence
Quarterly
Staleness state
Current
Question availability
Public benchmark set
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 AA Briefcase measure?
An independently evaluated professional-work benchmark reported as Elo.
Which model scores highest on AA Briefcase?
Claude Sonnet 5.5 by Anthropic currently leads with a score of 1811 on AA Briefcase.
How many models are evaluated on AA Briefcase?
23 AI models have published results on AA Briefcase in the BenchLM catalog.
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.