#GPT-6.1 Sol

89 stories · 13 in 7 days · 26 in 30 days
Stories, products and papers tagged "GPT-6.1 Sol", newest first.
Nov 13
ModelsUS vs ChinaMulti-source04:20
Chinese contentGoogle Gemini 4 Hydrogen 发布四周即超越自家更大的 Argon

Google 把 Flash 系列换成了 Gemini 4 Hydrogen,结果这个小模型上线四周就在 Artificial Analysis 榜单上超过了自家更贵的 Argon。它和 Claude Sonnet 5.5 得分差不多还更便宜,但比 GPT-6.1 Sol 贵 2.5 倍、写智能体代码不行。做编程或智能体的开发者可以对比一下再选。

Event
Official accountDecoder@Matthias BastianSource
ProductsBuildersMulti-source03:35
Chinese contentGemini 4 Hydrogen 已上线 Devin Desktop 和 Devin CLI

Cognition 把 Gemini 4 Hydrogen 接进了 Devin Desktop 和 Devin CLI,在自家 FrontierCode 2.0 基准上拿了 38.2%,只差 GPT-6.1 Sol 一名。平时用 Devin 写代码的人现在可以多一个模型选项,做模型选型的团队可以看看这份真实工程任务榜单再决定。

Event
Cognition@cognition_labsSource
ModelsUS vs ChinaMulti-source02:26
Chinese contentGemini 4 Hydrogen 成为 Agent Arena 速度最快的模型

做 Agent 应用的可以看看这份实测数据:Gemini 4 Hydrogen 单步只要 7.3 秒,10 分钟完成 82% 任务,速度比 Claude Sonnet 5.5 和 GPT-6.1 Sol 快不少,质量分数还没掉。对任务等待时间敏感的团队值得对比一下。

Event
lmarena.ai@lmarena_aiSource
ModelsUS vs ChinaMulti-source87°02:08
Chinese contentGoogle 发布 Gemini 4 Hydrogen,智能水平接近 Claude Sonnet 5.5 而成本仅其三成

Google 六周内第三个新模型,54 分智能打平 Sonnet 5.5 和 GPT-6.1 Sol,幻觉率才 8%,是 Claude 的六分之一,跑长文档视觉推理很强。做文档分析和知识工作流的人可以重点看,但搞终端编程智能体的还是 Sonnet 更稳。

Event
Official accountArtificial Analysis@ArtificialAnlysSource
Nov 12
ProductsBuildersMulti-source01:29
Chinese contentn8n Labs 开源个人智能体 Nod:在虚拟机里跑腿,密钥永不经手模型

n8n Labs 开源了一个叫 Nod 的个人智能体,能在自己电脑的虚拟机和浏览器里替你办事,没 API 的网站也能操作。亮点是账号密码和银行卡锁在 Vault 里,模型拿不到密钥,发钱删东西前都会先问你。做本地 Agent 的开发者值得看看这套安全设计。

Event
rohanpaul_ai@rohanpaul_aiSource
Nov 11
Nov 10
ModelsUS vs ChinaMulti-source18:29
Chinese content同一 3D 体素任务测试:Fable 5.2 花费 108.87 美元,GPT-6.1 Sol 仅 13.02 美元

有人拿同一个 3D 体素任务测了 Fable 5.2 和 GPT-6.1 Sol,成本差 8.4 倍但视频对比都放出来了。要挑视频生成或 3D 生成模型的话,这种带真实账单的横评比跑分更有参考价值。

Event
AI Will@FinanceYF5Source
Nov 9
Nov 8
Nov 6
Nov 5
Nov 3
Nov 2
Nov 1
Oct 28
Oct 27
ModelsBuildersMulti-sourceSelected06:23
Artificial Analysis launches Code2Video-AA video benchmark

The 300x spread in cost per task means the ranking alone is misleading: GPT-6.1 Sol delivers fourth place at 90% less cost than the leader, which changes which model is actually practical for batch video generation. Teams building automated marketing video pipelines can use the five-axis Elo breakdown to match models to their budget and quality needs, and the open-weights results give a concrete quality gap for self-hosting comparisons.

Event
Official accountArtificial Analysis@ArtificialAnlysSource
Oct 20
Oct 14
Oct 13
Oct 12
How-toBuildersMulti-sourceSelected19:39
BaoCut Integrates Reverse-Engineered Claude Design Skill

The workflow itself is the lesson: pairing a reverse-engineering tool like REA with a strong coding model can reproduce closed product features as open, inspectable skills, without ever copying the original prompts. Developers building agent tools can follow his five-step process with Codex and Claude Code 5.5, or just try the design skill in BaoCut locally. BaoCut is open source on GitHub, so the skill and frontend code are fully viewable.

Event
宝玉@doteySource
Oct 9
ProductsBuildersOfficialSelected03:12
OpenAI expands GPT-6.1 Sol access and EU data residency

Data residency options determine whether regulated industries in Europe can deploy a model at all, so the addition of EU residency for these tiers removes a common procurement blocker. Teams on Pro 500 or Enterprise plans should check with their admins to enable the new access tiers. Compared with regional offerings, residency coverage is often the deciding factor between providers for compliance-sensitive workloads.

Event
Primary source@OpenAIDevs@OpenAIDevsSource