Pipette係Liquid AI推出嘅免費開源 on device AI benchmark,用完整部署組合,而唔係單一模型作為測量單位。 平台將模型、量化格式、runtime、裝置同工作負載一併評估,公開數據集涵蓋超過1,000個經實驗室驗證嘅配置,並提供五項裝置端效能指標。[11][4] 目前支援iOS及Android,測試時要先將模型下載到裝置本機;涵蓋Liquid自家LFM系列,以及Gemma、Nanbeige等第三方細型模型。[10][8]
手機 AI 智能評測以16K tokens作上下文限制;Nanbeige4.2 3B同LFM2.5 2.6B喺平均分同得63分。[9][8]
What is Liquid AI’s Pipette, the free open source on device AI benchmarking platform released on August 24 by the startup founded by formerAI-generated editorial hero image for What is Liquid AI’s Pipette, the free open source on device AI benchmarking platform released on August 24 by the startup founded by former.
AI 提示
Create a landscape editorial hero image for this Studio Global article: What is Liquid AI’s Pipette, the free open source on device AI benchmarking platform released on August 24 by the startup founded by former. Article summary: Pipette is Liquid AI’s free, open source benchmark suite for measuring an actual on device deployment—not merely a model—across the combination of model, quantization, runtime, device, and workload.. Topic tags: general web, llm, agents, ai, workflow. Style: premium digital editorial illustration, source-backed research mood, clean composition, high detail, modern web publication hero. Use reference image context only for broad subject, composition, and topical grounding; do not copy the exact image. Avoid: logos, brand marks, copyrighted characters, real person likenesses, fake screenshots, UI text, readable text, watermarks, charts with fake numbers, clickbait
openai.com
Pipette係Liquid AI嘅免費開源 benchmark suite,專門量度 AI 模型喺手機、電腦等 edge device 上面真正部署之後嘅表現。佢測試嘅唔係「模型本身」或者一個孤立嘅 tokens per second 數字,而係以下完整組合:模型、量化格式、runtime、裝置,以及工作負載。119
即係話,現時iPhone同Android嘅 throughput 數字,唔係純粹比較兩粒手機晶片。Apple一邊可能包括GPU加速,Android一邊就係CPU推理。Android結果仍然可以用嚟了解CPU效能,以及喺該backend下嘅實際使用體驗,但唔足以證明某部手機整體本地 AI 硬件一定快過另一部。104
如果日後加入Android GPU及NPU等對等加速backend,跨平台比較先會更加有意義。
同手機晶片 AI 競賽有咩關係?
Pipette推出之際,晶片廠商正集中強調本地及 agentic AI 工作負載。高通表示,Snapdragon 8 Elite Gen 5嘅第三代Oryon CPU最高時脈達4.74GHz,並聲稱該平台CPU效能提升20%、CPU能效提升35%。14
另外,高通亦預告後續旗艦Snapdragon平台,Oryon CPU目標時脈達5GHz,並加入FlexCache設計,讓不同核心動態共享cache。呢啲公布反映業界正追求更快、更即時嘅本地 AI 體驗,但單靠時脈高低,並唔能夠預測大型語言模型嘅實際推理速度。913
Pipette係Liquid AI推出嘅免費開源 on device AI benchmark,用完整部署組合,而唔係單一模型作為測量單位。 平台將模型、量化格式、runtime、裝置同工作負載一併評估,公開數據集涵蓋超過1,000個經實驗室驗證嘅配置,並提供五項裝置端效能指標。[11][4] 目前支援iOS及Android,測試時要先將模型下載到裝置本機;涵蓋Liquid自家LFM系列,以及Gemma、Nanbeige等第三方細型模型。[10][8]
手機 AI 智能評測以16K tokens作上下文限制;Nanbeige4.2 3B同LFM2.5 2.6B喺平均分同得63分。[9][8]
What is Liquid AI’s Pipette, the free open source on device AI benchmarking platform released on August 24 by the startup founded by formerAI-generated editorial hero image for What is Liquid AI’s Pipette, the free open source on device AI benchmarking platform released on August 24 by the startup founded by former.
AI 提示
Create a landscape editorial hero image for this Studio Global article: What is Liquid AI’s Pipette, the free open source on device AI benchmarking platform released on August 24 by the startup founded by former. Article summary: Pipette is Liquid AI’s free, open source benchmark suite for measuring an actual on device deployment—not merely a model—across the combination of model, quantization, runtime, device, and workload.. Topic tags: general web, llm, agents, ai, workflow. Style: premium digital editorial illustration, source-backed research mood, clean composition, high detail, modern web publication hero. Use reference image context only for broad subject, composition, and topical grounding; do not copy the exact image. Avoid: logos, brand marks, copyrighted characters, real person likenesses, fake screenshots, UI text, readable text, watermarks, charts with fake numbers, clickbait
openai.com
Pipette係Liquid AI嘅免費開源 benchmark suite,專門量度 AI 模型喺手機、電腦等 edge device 上面真正部署之後嘅表現。佢測試嘅唔係「模型本身」或者一個孤立嘅 tokens per second 數字,而係以下完整組合:模型、量化格式、runtime、裝置,以及工作負載。119
即係話,現時iPhone同Android嘅 throughput 數字,唔係純粹比較兩粒手機晶片。Apple一邊可能包括GPU加速,Android一邊就係CPU推理。Android結果仍然可以用嚟了解CPU效能,以及喺該backend下嘅實際使用體驗,但唔足以證明某部手機整體本地 AI 硬件一定快過另一部。104
如果日後加入Android GPU及NPU等對等加速backend,跨平台比較先會更加有意義。
同手機晶片 AI 競賽有咩關係?
Pipette推出之際,晶片廠商正集中強調本地及 agentic AI 工作負載。高通表示,Snapdragon 8 Elite Gen 5嘅第三代Oryon CPU最高時脈達4.74GHz,並聲稱該平台CPU效能提升20%、CPU能效提升35%。14
另外,高通亦預告後續旗艦Snapdragon平台,Oryon CPU目標時脈達5GHz,並加入FlexCache設計,讓不同核心動態共享cache。呢啲公布反映業界正追求更快、更即時嘅本地 AI 體驗,但單靠時脈高低,並唔能夠預測大型語言模型嘅實際推理速度。913
Artificial Analysis has published the results of its measurements of the performance of various AIs on iPhones, measuring performance in a realistic environment by limiting context length and response time.
Artificial Analysis has published the results of its measurements of the performance of various AIs on iPhones, measuring performance in a realistic environment by limiting context length and response time.