#评测/基准
Artificial Analysis@ArtificialAnlysAI 评分4646
Artificial Analysis@ArtificialAnlysAI 评分5454
OpenRouter@OpenRouterAI 评分3838小贴士:看看生态中不同路由器如何使用 https://openrouter.ai/benchmarks/routers 进行基准测试 不同路由器相对前沿模型的成本对比 👇
Artificial Analysis@ArtificialAnlysAI 评分4848
Artificial Analysis@ArtificialAnlysAI 评分4545
meng shao@shao__mengAI 评分6666引用Hamel Husain@HamelHusainNew blog post 🎉: How To Use Decision Models For Evals In this post, we show you how to choose a decision model for your task, tune its threshold (don't assume the 0.5 default is good!), and interpret its scores. Lots of cool interactive charts here. https://x.com/i/article/2112552453422104576
IT之家·人工智能AI 评分6767 OpenAI 报告称 GPT-6.1 Sol 的 COBOL 迁移准确率最高达 97.9%
OpenAI 于 10 月 20 日发布研究报告,介绍其增强 GPT-6 Astra 等模型构建和迁移 COBOL、Fortran、Visual Basic 等低资源编程语言的能力,并推出 MultiLangBench(20 种语言从零构建兼容 SQLite 的数据库引擎)和 COBOL 迁移至 C++、Java、Python、Rust 两项评测。
Krea@krea_aiAI 评分3939Krea 3 在电商领域无可匹敌。 像素级完美的产品一致性与闪电般的速度。 https://krea.ai/krea-3
引用Krea@krea_aitoday, Krea 3 makes it to the top of image arenas. - "Fast" pushes the pareto curve of speed and quality. - "Ultra" ranks top-4, only surpassed by OpenAI models. the only company valued under $1T in the top-10! more below 👇
Dongxi 东锡 NLP@dongxi_nlpAI 评分6464引用Hamel Husain@HamelHusainNew blog post 🎉: How To Use Decision Models For Evals In this post, we show you how to choose a decision model for your task, tune its threshold (don't assume the 0.5 default is good!), and interpret its scores. Lots of cool interactive charts here. https://x.com/i/article/2112552453422104576
MiniMax (official)@MiniMax_AIAI 评分4040感谢 Parallel Web System 提供如此有意义的数据点~ 是的,MiniMax M3.1 Flash 现已在搜索能力排行榜的效率维度上位列第一。而且它很快~
引用Parallel Web Systems@p0MiniMax M3.1 Flash is now #1 on our search capabilities leaderboard for efficiency. It delivers web research performance with Parallel Search similar to DeepSeek V4.1 Flash (1017) at ~3x lower cost.
Arena.ai@arenaAI 评分6060引用Arena.ai@arenaExciting news: Muse Wonder Full (Max) by @AIatMeta has just debuted in the Code Arena: WebDev at #6 overall with 1726 pts, putting Meta in the top 3 labs! This is +75 pts and 11 places above Muse Wonder (Max) at #17 (1651 pts), and +67 pts and 8 places above Muse Spark 1.3 (Max) at #14 (1659 pts). Muse Wonder Full (#6) also landed +16 pts above Claude Sonnet 5.5 (#17), and +48 pts above Gemini 4 Argon (#10). With this score, Muse took the #3 lab spot (previously held by Google). Across categories, Muse Wonder Full (Max) landed: - #4 in Data & Analytics and in Consumer Product - #5 in Brand & Marketing - #6 in Content Creation Tools - #7 in Gaming and in Simulations - #8 in Reference-Based Design Congrats to the @AIatMeta team and @alexandr_wang on this release!
Aravind Srinivas@AravSrinivasAI 评分3838Perplexity Decider 在 Vals AI 的 Financial ClaimBench 上达到帕累托最优
引用Vals AI@ValsAIToday we’re releasing ClaimBench, our new benchmark targeted towards decision models. It tests whether a model can tell if a financial document supports, contradicts, or doesn’t address a claim. We evaluated 15 models on 1,960 claims paired with 10-K excerpts, with Decider V1.2 leading the pack of decision models.
Arena.ai@arenaAI 评分5757引用Arena.ai@arenaUpdate to the Agent Arena Pareto frontier: GPT-6 Luna (Max) has landed on the Agent Arena Pareto frontier with +1.59% net improvement at a median cost of just $0.05 per task! Luna costs 29% less than DeepSeek-V4.1-Flash and 76% less than Hy4 preview, while trailing their net-improvement scores by just 2.27 and 2.51 percentage points respectively. Compared to other GPT models on the Pareto frontier, GPT-6 Luna (Max) costs 94% less than GPT-6 Sol (Max), and 98% less than GPT-6 Astra (Max). Congrats to the @OpenAI team on this cost-efficient release!
fal@fal精选AI 评分6868引用fal@fal隆重推出 FLUX 3 Max,这是 fal Research 与 @bfl_ai 合作推出的最新后训练视频模型。 在独立第三方基准测试中,FLUX 3 Max 在图生视频质量上并列第一,可在约 12 秒内生成一段 10 秒、720/768p 的视频。 在接下来的三周内,FLUX 3 Max 享受 75% 折扣,使其成为同等质量水平下最实惠的前沿视频模型。
推荐理由:原文给出第三方 Elo 分数和生成速度数据,读者可据此判断该模型在图生视频梯队中的位置。
The Decoder:AI NewsAI 评分6161 OpenAI 称 GPT-6 在 COBOL、Fortran 等遗留编程语言上大幅提升
OpenAI 表示最新模型针对 COBOL、Fortran、Visual Basic 等低资源语言做了改进,这些语言训练代码少但支撑大量关键基础设施,IBM 估计 COBOL 支撑超过 40% 的网上银行系统和 95% 的 ATM 交易。
Artificial Analysis@ArtificialAnlysAI 评分6464Artificial Analysis 分析 Intelligence Index 各阈值达成的成本变化:达到 30 分如今每任务 0.03 美元(GPT-6 Luna high)。

Artificial Analysis@ArtificialAnlysAI 评分6060Artificial Analysis 评测了 Mistral 发布的 Voxtral 3 语音模型,包括 1 个 TTS 模型和 3 个语音转文本模型。
Hugging Face 社区博客AI 评分6060 Ai2 发布 FairStress 基准,测出 LLM 决策中的反刻板印象过度纠正
Ai2 发布 FairStress 基准,用 300 个场景扩展成超过 1300 万测试用例,测量大语言模型在高风险决策中面对对抗前缀时的公平性。评测发现模型在身份信息显式时更易放弃支持少数群体编码候选人的正确决策,平均放弃差距为 6.9(隐含身份)至 9.3 个百分点(显式标签)。基准已开放下载并集成到 Olmo-eval,测试的九种去偏方法中仅反事实链式推理在保留准确率的同时缩小差距。
Artificial Analysis@ArtificialAnlysAI 评分5757
Artificial Analysis@ArtificialAnlysAI 评分5959
Artificial Analysis@ArtificialAnlysAI 评分6363Artificial Analysis 评测了 Mistral 新发布的 Voxtral 3 系列语音模型,包括 1 个 TTS 和 3 个语音转文字模型。
Arena.ai@arenaAI 评分5151Arena 宣布 Krea 3 Ultra 登上 Single-Image Edit Arena 和 Multi-Image Edit Arena,Krea 跻身两个榜单按实验室计前 5。
引用Krea@krea_aitoday, Krea 3 makes it to the top of image arenas. - "Fast" pushes the pareto curve of speed and quality. - "Ultra" ranks top-4, only surpassed by OpenAI models. the only company valued under $1T in the top-10! more below 👇
Ai2 / Allen Institute for AI精选AI 评分6161 Ai2 发布 FairStress 基准:测量 LLM 决策中被隐藏的公平性不对称
Ai2 发布 FairStress 基准,通过 300 个两候选决策场景和对抗前缀,测量 LLM 在面对说服性论据时改变决策的难易是否因候选人所属人口群体而不同,全部组合超过 1300 万测试用例,并已开放下载、集成进 Olmo-eval。
推荐理由:Ai2 用对抗前缀测出模型在人口群体之间的决策不对称,并发现多数去偏方法不可靠,提供了诊断与改进思路。
🚨 AI News | TestingCatalog@testingcatalogAI 评分4949
引用Krea@krea_aitoday, Krea 3 makes it to the top of image arenas. - "Fast" pushes the pareto curve of speed and quality. - "Ultra" ranks top-4, only surpassed by OpenAI models. the only company valued under $1T in the top-10! more below 👇
Nathan Lambert@natolambertAI 评分3030防御性网络安全方面可能也很出色,但没人会发这种内容来博眼球!
引用Vals AI@ValsAIWe found DeepSeek V4.1 Flash 1017 beats GPT 6 Sol on our multi-step offensive cybersecurity task. Sol was tested via the Daybreak trusted access program. We have been building a cybersecurity benchmark based on simulated corporate networks. This is a valuable indicator for the cyber-risk a model poses if in the hands of a bad actor. We plan to make the full evaluation public in the coming weeks but wanted to share our first immediate finding here.
Artificial Analysis@ArtificialAnlysAI 评分6060Artificial Analysis 盘点上周美国开源权重模型在智能和参数效率上的进展。

IT之家·人工智能AI 评分6868 DeepSeek-V4.1-Flash-1017 登顶 Agent Arena 与 Code Arena: WebDev 开源榜,总榜第 11
Arena 宣布 DeepSeek-V4.1-Flash-20261017 (Max) 登顶 Agent Arena 和 Code Arena: WebDev 开源模型榜单,总榜排名第 11。
Artificial Analysis 完整文章AI 评分5353 Artificial Analysis 发布 HealthBench 与 HealthBench Professional 模型健康对话排行榜
Artificial Analysis 上线 HealthBench 和 HealthBench Professional 排行榜,两个基准来自 OpenAI,用医生编写的评分标准评估模型在真实健康对话中的表现。
Hamel Husain 长文AI 评分6666 Hamel Husain 详解如何用决策模型做自动评测
Hamel Husain 用 100 条公寓租赁客服 trace 实测 Jev 1.13、Perplexity Decider 1.1 27B、Liquid d1 与 d1-ultra、Cloudflare Clef Flash 和 OpenAI Decisions(GPT-6 Luna)识别失败案例。
Artificial Analysis@ArtificialAnlysAI 评分6060
Stanford HAI News精选AI 评分6767 斯坦福 HAI 发布 MedHELM、PhysicianBench 与 HealthAdminBench,测出医疗 AI Agent 实际完成率不足一半
斯坦福团队总结三年构建医疗 AI 基准的成果,PhysicianBench 上最佳模型 GPT-5.5 单次仅完成 46% 任务,HealthAdminBench 上 Claude Opus 4.6 完成 36.3%。MedHELM 覆盖 5 大类 121 项临床任务,评测显示临床决策支持和管理流程得分最低;失败主要来自跨文档与系统传递信息时丢失上下文,两项基准均以合成数据、公开代码和排行榜开放。
推荐理由:斯坦福团队基于三年基准构建经验给出实测结论,最好的医疗 Agent 完成任务不足一半,失败多源于跨系统信息传递。