摘要
AA-Omniscience v1.1 Non-Hallucination Rate
规则生成永久链接
https://daily.maas.click/item/obs_38934f3779ba3f51db3ca03d3bfb4b943253c499a87c67c76f65a8f52222ac9f/变化摘录
以下为已保存的变化摘录
证据级别：来源差异（source_diff）——这是抓取到的来源页面变化，其中定价页文本变化不代表已验证的调价。
新增 24 行
- AA-Omniscience v1.1 Accuracy
- AA-Omniscience v1.1 Non-Hallucination Rate
- Announcing Artificial Analysis Intelligence Index v5
- Artificial Analysis Intelligence Index v5.0
- Artificial Analysis Intelligence Index v5.0 incorporates 9 evaluations: AA-Briefcase v1.1, AutomationBench-AA, Terminal-Bench 4.0, FrontierCode 2.0, AA-Omniscience v1.1, GDP.pdf, HLE-Diamond, CritPt v2 (Hard Subset), and Terminal-Bench-Science 0.1
- Artificial Analysis Progress Index v1, Over Time
- CritPt v2 (Hard Subset)
- Each lab’s line shows its highest-scoring model available at that point in time, using each model’s highest-scoring effort variant. Read the Progress Index v1 methodology.
- For example, an older model might have been tested on easier benchmarks and a newer model on harder ones. Models tested on both sets help us estimate the difference in difficulty and compare the older and newer models on the same scale.
- FrontierCode 2.0
- Gemini 4 Carbon (High)
- Gemini 4 Carbon is the new #1 on the Artificial Analysis Intelligence Index v5
- GitHub Copilot Fusion orchestrates Anthropic and OpenAI models to balance performance and cost
- GoogleAnthropicOpenAIMetaKimiDeepSeekAlibabaSpaceXAIZ AIInclusionAIXiaomiStepFunMiniMaxReflectionInstitute of Foundation ModelsThinking MachinesMistral
- GoogleAnthropicOpenAIMetaKimiSpaceXAIDeepSeekZ AIXiaomiOther Models
- HLE-Diamond
- HealthBench and HealthBench Professional: Evaluating models in realistic health conversationsSee more
- How the Progress Index v1 is calculated
- Intelligence Index v5
- Intelligence IndexUpdatedOpen Weights / ProprietaryText Only / Multimodal InputsBy Country
- Item response theory estimate of general capability across Artificial Analysis evaluations · Highest-scoring effort variant shown
- Methodology updated · 27 Oct
- Mistral Large 4 (High)
- Mistral Large 4 (Medium)
删除 5 行
- AA-LCR v1.1
- AA-Omniscience Accuracy
- AA-Omniscience Non-Hallucination Rate
- Announcing AA-3D: a benchmark for interactive 3D experiences
- AnthropicMetaKimiOpenAIGoogleSpaceXAIXiaomiAlibabaDeepSeekZ AIInclusionAIStepFunMiniMaxMistralInstitute of Foundation ModelsReflectionThinking Machines
修订历史（2 版）
- r1
- r2