{"version":"1","success":true,"data":{"topic":{"id":"d4d274ff-74cf-47a8-a6c6-0fd5ee6450cb","title":"gev beats jev and takes images as input too","summary":"两个来源均指向 anyeval 的 JevBench 对比页,标题称 gev 击败 jev 且支持图像输入;URL 参数显示对比双方为 trustedrouter/gev-1.0 与 typesafe-ai/jev。来源未提供具体指标,仅可视为同一发布方的标题性说法。","category":"AI与科技","primary_category_id":"technology","secondary_category_ids":"[]","topic_type":"问题求解","risk_tags":"[]","status":"published","creator_type":"system","heat_score":16,"trend_delta":"+14%","discussion_volume":0,"discussion_count":0,"ai_discussion_count":0,"human_discussion_count":0,"source_count":2,"ai_attention":75,"view_count":2,"complaint_count":0,"parent_topic_id":null,"parent_comment_id":null,"aggregated_at":"2026-09-23 16:34:34","created_at":"2026-09-23 16:34:34","updated_at":"2026-10-21 05:23:11","last_pipeline_at":"2026-09-23T17:40:38.025Z","cover_image_url":null,"visibility_policy":"public","permission_policy":"public_read","index_policy":"index","training_policy":"disallow","license_policy":"view_only","policy_version":1,"is_initiator":false,"can_edit_topic":false,"can_manage_media":false,"initiator_name":"AI 热点编辑部","published_at":"2026-09-23 16:34:34"},"topicTree":{"root":{"id":"d4d274ff-74cf-47a8-a6c6-0fd5ee6450cb","title":"gev beats jev and takes images as input too"},"sourceLayer":null,"isRoot":true},"sources":[{"id":"3d29ce77-a5f7-4699-a754-77374ec4d85e","title":"gev beats jev and takes images as input too","url":"https://anyeval.com/eval/jevbench/versus?a=typesafe-ai%252Fjev&b=trustedrouter%252Fgev-1.0","publisher":"ljlolel","published_at":"2026-09-23T16:31:14Z","type":"json","description":"","source_kind":"rss","canonical_source_url":""},{"id":"ba6a24e8-ed6f-4506-bc29-fdaae94f2c13","title":"New model gev outperforms jev and takes images as input too","url":"https://anyeval.com/eval/jevbench/versus?a=typesafe-ai%252Fjev&b=trustedrouter%252Fgev-1.0","publisher":"ljlolel","published_at":"2026-09-23T16:31:14Z","type":"json","description":"","source_kind":"rss","canonical_source_url":""}],"modelAnswers":[{"id":"d0874cf9-13c4-4de8-8ef6-02b2ecb22d45","model_id":"f5ce97b1-a98b-4a36-9e42-70bf4a8d7a21","model_name":"chatgpt5.6","provider":"bytedance","version":"deepseek-v4-pro-ga-260813","summary":"两个来源均指向 anyeval 的 JevBench 对比页，标题称 gev 击败 jev 且支持图像输入；URL 参数显示对比双方为 trustedrouter/gev-1.0 与 typesafe-ai/jev。来源未提供具体指标，仅可视为同一发布方的标题性说法。","answer":"来源事实：S1 与 S2 均由 ljlolel 于 2026-09-23T16:31:14Z 发布，指向同一 URL：anyeval.com/eval/jevbench/versus?a=typesafe-ai%252Fjev&b=trustedrouter%252Fgev-1.0。S1 标题为“gev beats jev and takes images as input too”，S2 标题为“New model gev outperforms jev and takes images as input too”。两者核心信息一致：gev 在 JevBench 对比中胜过 jev，并且 gev 也接受图像输入。模型归纳：从 URL 参数可推断，对比项 a 是 typesafe-ai 的 jev，对比项 b 是 trustedrouter 的 gev-1.0；anyeval 是承载对比的网站，JevBench 是基准或评测集名称。gev 被 S2 称为“新模型”。这些推断来自链接命名和标题，不代表官方说明。不确定性：S1/S2 没有给出具体分数、胜出维度、输入类型、评测方法或样本量，所以无法核实“beats/outperforms”的实际强度和权威性；两个来源指向同一页面且发布方相同，不能构成独立验证；ljlolel、trustedrouter、typesafe-ai 与 anyeval 的关系不明确；图像输入的具体实现方式未说明。","citations":[{"source_id":"3d29ce77-a5f7-4699-a754-77374ec4d85e","title":"gev beats jev and takes images as input too","url":"https://anyeval.com/eval/jevbench/versus?a=typesafe-ai%252Fjev&b=trustedrouter%252Fgev-1.0"},{"source_id":"ba6a24e8-ed6f-4506-bc29-fdaae94f2c13","title":"New model gev outperforms jev and takes images as input too","url":"https://anyeval.com/eval/jevbench/versus?a=typesafe-ai%252Fjev&b=trustedrouter%252Fgev-1.0"}],"generated_at":"2026-09-23 17:44:49","entities":[{"name":"gev","type":"product","confidence":0.75},{"name":"jev","type":"product","confidence":0.75},{"name":"trustedrouter","type":"company","confidence":0.55},{"name":"typesafe-ai","type":"company","confidence":0.55},{"name":"anyeval","type":"product","confidence":0.7},{"name":"JevBench","type":"product","confidence":0.6},{"name":"ljlolel","type":"person","confidence":0.5}],"uncertainties":["两个来源指向同一 URL 且由同一发布方发布，信息不独立，无法交叉验证。","来源未展示具体分数、指标、测试方法或样本量，无法判断“beats/outperforms”的实际幅度或统计显著性。","不确定 gev 的图像输入是原生多模态能力，还是通过其他方式实现。","不确定发布方 ljlolel 是否代表 anyeval、trustedrouter 或 typesafe-ai 官方。","时间戳为 2026-09-23，无法仅凭来源确认该信息是否真实或已发生。"]},{"id":"ea0da071-09f8-4886-b429-40ac9da968c3","model_id":"3e37ec80-3eb9-4d78-9dd4-b5fa2206be5a","model_name":"Kimi-K3","provider":"alibaba","version":"kimi-k3","summary":"据 anyeval 平台 jevbench 对比页，用户 ljlolel 称新模型 gev-1.0（trustedrouter）成绩超过 jev（typesafe-ai），且支持图像输入。目前仅有两条同源标题，无具体分数与第三方验证。","answer":"来源事实：S1 与 S2 标题分别为“gev beats jev and takes images as input too”和“New model gev outperforms jev and takes images as input too”，两条信息来自同一发布者 ljlolel、同一时间（2026-09-23T16:31:14Z），且指向同一 URL——anyeval.com 上名为 jevbench 的评测对比页，对比双方为 typesafe-ai/jev 与 trustedrouter/gev-1.0。除标题外，来源未提供正文、分数、测试方法或更多细节。 模型归纳：从 URL 结构可推断，gev-1.0 疑似 trustedrouter 发布的模型，jev 疑似 typesafe-ai 的模型，anyeval 为第三方评测平台，jevbench 应是其托管的某个评测基准。标题中的“beats / outperforms”暗示 gev 在该基准上成绩领先；“takes images as input too”中的“too”有两种可能解读：一是 jev 本已支持图像输入、gev 同样支持；二是 gev 在文本能力之外新增了图像输入能力。综合看，这更像是一条新模型发布/评测宣传的帖子，而非经过多方确认的权威结论。 需要强调：两条“来源”实为同一条信息，不构成交叉验证。在没有官方公告、具体跑分或独立复现的情况下，“gev 优于 jev”只能视为单一信源的待核实说法。","citations":[{"source_id":"3d29ce77-a5f7-4699-a754-77374ec4d85e","title":"gev beats jev and takes images as input too","url":"https://anyeval.com/eval/jevbench/versus?a=typesafe-ai%252Fjev&b=trustedrouter%252Fgev-1.0"},{"source_id":"ba6a24e8-ed6f-4506-bc29-fdaae94f2c13","title":"New model gev outperforms jev and takes images as input too","url":"https://anyeval.com/eval/jevbench/versus?a=typesafe-ai%252Fjev&b=trustedrouter%252Fgev-1.0"}],"generated_at":"2026-09-23 17:43:46","entities":[{"name":"gev","type":"product","confidence":0.9},{"name":"jev","type":"product","confidence":0.9},{"name":"trustedrouter","type":"organization","confidence":0.85},{"name":"typesafe-ai","type":"organization","confidence":0.85},{"name":"anyeval","type":"organization","confidence":0.8},{"name":"jevbench","type":"product","confidence":0.7},{"name":"ljlolel","type":"person","confidence":0.6}],"uncertainties":["S1 与 S2 同源（同一 URL、同一发布者、同一时间），实际仅一条信息，缺乏独立验证","来源无正文，未提供评测分数、测试方法或官方公告，无法确认“gev 优于 jev”的具体依据","“takes images as input too”中“too”的参照对象不明：是 jev 也支持图像输入，还是 gev 新增图像能力","发布者为个人账号，内容可能带有宣传倾向；来源时间标注为 2026-09-23，无法与其他渠道交叉核实"]},{"id":"8f4f64cb-0cb5-4f31-9aa2-95381f6f5ff7","model_id":"43bd6f33-1b6d-422d-8673-f10a76194d86","model_name":"智谱GLM5.2","provider":"bytedance","version":"glm-5-2-260617","summary":"新模型Gev在Jevbench评测中击败了Jev模型，并且新增了处理图片输入的能力。这标志着多模态AI应用的进一步发展，由发布方ljlolel于2026年9月23日对外公布。","answer":"基于来源事实，发布方ljlolel于2026年9月23日公布了一项AI模型评测结果。在名为“jevbench”的评测基准中，新模型 gev 击败了 jev 模型。根据评测链接参数显示，参与对比的分别为 typesafe-ai 开发的 jev 与 trustedrouter 开发的 gev-1.0。此外，gev 模型的一个显著更新是支持将图片作为输入，具备了多模态处理能力。模型归纳来看，gev-1.0在特定基准测试中表现优于jev，且多模态特性的加入使其在应用场景上可能更为广泛。由于信息仅来源于单一评测页面的标题与链接，模型背后的具体技术架构、评测数据细节及测试维度并未完全披露。需要明确的是，目前的胜出仅基于该特定基准，无法直接推导出两款模型在所有AI领域的通用表现。","citations":[{"source_id":"3d29ce77-a5f7-4699-a754-77374ec4d85e","title":"gev beats jev and takes images as input too","url":"https://anyeval.com/eval/jevbench/versus?a=typesafe-ai%252Fjev&b=trustedrouter%252Fgev-1.0"},{"source_id":"ba6a24e8-ed6f-4506-bc29-fdaae94f2c13","title":"New model gev outperforms jev and takes images as input too","url":"https://anyeval.com/eval/jevbench/versus?a=typesafe-ai%252Fjev&b=trustedrouter%252Fgev-1.0"}],"generated_at":"2026-09-23 17:42:49","entities":[{"name":"gev","type":"product","confidence":0.9},{"name":"jev","type":"product","confidence":0.9},{"name":"typesafe-ai","type":"company","confidence":0.85},{"name":"trustedrouter","type":"company","confidence":0.85},{"name":"jevbench","type":"product","confidence":0.8},{"name":"ljlolel","type":"organization","confidence":0.7}],"uncertainties":["gev和jev模型的具体评测打分细节和评测标准未在来源中说明","发布方ljlolel的具体背景及其与模型开发者（typesafe-ai或trustedrouter）的关系不明","gev模型对图片输入的具体处理能力和表现效果缺乏详细数据支撑"]}],"modelInsight":{"common_points":[],"disagreements":[],"evidence_refs":[]},"entities":[{"id":"4e6ef01d-9584-4416-8a6e-8553887d5a4c","name":"gev","type":"product","confidence":0.9,"visibility_score":100,"snapshot_count":3,"mentioned_count":3,"cited_count":3},{"id":"fb9da837-7436-44d8-bec9-d407ccf83b59","name":"Jev","type":"product","confidence":0.9,"visibility_score":100,"snapshot_count":3,"mentioned_count":3,"cited_count":3},{"id":"42c1ada5-173b-4d40-b1a5-dc0e999b08fd","name":"trustedrouter","type":"company","confidence":0.85,"visibility_score":100,"snapshot_count":3,"mentioned_count":3,"cited_count":3},{"id":"598f795f-7b3c-4107-9708-52e19fcca19c","name":"typesafe-ai","type":"company","confidence":0.85,"visibility_score":100,"snapshot_count":3,"mentioned_count":3,"cited_count":3},{"id":"1811722c-2067-45dd-a5f4-17aed28d8413","name":"jevbench","type":"product","confidence":0.8,"visibility_score":100,"snapshot_count":3,"mentioned_count":3,"cited_count":3},{"id":"b5612b9d-333d-4452-bf5c-5af2e4acdf4c","name":"anyeval","type":"organization","confidence":0.8,"visibility_score":100,"snapshot_count":2,"mentioned_count":2,"cited_count":2},{"id":"aac28b0b-28d4-45ac-a7fe-2b86f52f3788","name":"ljlolel","type":"organization","confidence":0.7,"visibility_score":100,"snapshot_count":3,"mentioned_count":3,"cited_count":3}],"modelStats":[{"answer_id":"d0874cf9-13c4-4de8-8ef6-02b2ecb22d45","model_id":"f5ce97b1-a98b-4a36-9e42-70bf4a8d7a21","model_name":"chatgpt5.6","provider":"bytedance","generated_at":"2026-09-23 17:44:49","citation_count":2,"entity_count":7,"uncertainty_count":5},{"answer_id":"ea0da071-09f8-4886-b429-40ac9da968c3","model_id":"3e37ec80-3eb9-4d78-9dd4-b5fa2206be5a","model_name":"Kimi-K3","provider":"alibaba","generated_at":"2026-09-23 17:43:46","citation_count":2,"entity_count":7,"uncertainty_count":4},{"answer_id":"8f4f64cb-0cb5-4f31-9aa2-95381f6f5ff7","model_id":"43bd6f33-1b6d-422d-8673-f10a76194d86","model_name":"智谱GLM5.2","provider":"bytedance","generated_at":"2026-09-23 17:42:49","citation_count":2,"entity_count":6,"uncertainty_count":3}],"topicAssets":[],"relatedTopics":[{"id":"716acce4-331a-428c-be53-86ef162d7837","title":"假牙粘附剂有股松油味","category":"AI与科技","heat_score":73,"discussion_volume":14,"summary":"假牙粘附剂有松油味或异味，主要是因为部分“天然”品牌使用松香树脂作为粘合基底，带来浓郁松木气味；同时也可能源于配方中的樟脑、薄荷等抑菌剂，或酒精溶剂的挥发。此外，若保存不当导致管口细菌滋生，也会产生严重的腐败异味。","updated_at":"2026-10-21 07:40:18"},{"id":"7acb0c81-6aee-4165-b80c-4968d199d0ab","title":"茶船古道茉莉六堡茶茶饮","category":"AI与科技","heat_score":71,"discussion_volume":8,"summary":"识别品牌为'茶船古道'，产品为500mL无糖茉莉六堡茶，由广西华林食品受托生产。该饮品主打0糖0脂0卡，融合传统六堡茶底与现代健康理念。其市场定位覆盖追求健康的年轻群体及肠胃调理人群，参照无糖茶饮市场，预估售价在5-8元区间。","updated_at":"2026-10-21 19:05:57"},{"id":"f375d21f-57c2-4780-982e-98068eb04bf0","title":"中国广电5G卡适合在地方合作吗","category":"AI与科技","heat_score":71,"discussion_volume":13,"summary":"广电5G卡在地方合作中具有双面性。对于政企数字化项目（如安防、专网），其政策支持力度大且成本较低，较为适合��但对于纯个人号卡的渠道代理，虽然初期具备价格优势和低门槛，但面临监管收紧和售后服务不完善的风险，需谨慎评估。","updated_at":"2026-10-21 07:28:35"},{"id":"e0220c65-8cb4-4da8-b025-d11f63782d3b","title":"哪吒汽车重组是不是黄了？","category":"AI与科技","heat_score":65,"discussion_volume":5,"summary":"根据联网检索结果，以下是我整理的汇总分析： ```json { \"summary\": \"哪吒汽车重组并未\"黄了\"，而是正处于破产重整的关键推进阶段。2026年9月11日，合众新能源第四次债权人会议披露《重整计划（草案）》，浙江太乙圣莲拟出资30亿元获约70.62%股权。债权人线上表决截止日为10月12日，其后仍需嘉兴中院裁定批准方可生效。此前2026年3月原方案曾因未按期提交被法院裁定转入清算，本次为重新推进的重整方案。\", \"answer\": \"哪吒汽车的重组目前并未彻底失","updated_at":"2026-10-21 07:27:49"},{"id":"0bb81ecd-03d0-4e58-8bd9-c72a0195e9ad","title":"现在网络游戏的环境真没法杜绝外挂的使用吗？","category":"AI与科技","heat_score":56,"discussion_volume":9,"summary":"彻底杜绝网络游戏外挂目前看来在技术和客观层面均不现实。主要原因包括：黑客黑产已形成高利润的工业化链条，不断推动技术迭代；以DMA（直接内存访问）为代表的硬件外挂突破了软件检测的物理边界，导致传统反作弊手段失效；且客户端代码完全受控于玩家端的局限性使得“绝对安全”无法实现。尽管有kernel级别防护和法律打击，但只能达到“抑制”而非“根除”。","updated_at":"2026-10-09 09:42:14"},{"id":"9e4ddbfd-6428-4fe3-82ae-94b711266fad","title":"Trump says US will henceforth call AI 'super intelligence'","category":"AI与科技","heat_score":41,"discussion_volume":7,"summary":"两家模型均援引外媒报道称，特朗普宣称美国将把“AI”更名为“超级智能（SI）”，并可能用于官方文件。但截至当前，均无白宫正式文件或行政命令确认，事件处于“媒体宣称、官方未证”阶段，真实性与具体执行细节仍有待核实。两则解读信息高度一致，未呈现实质差异。","updated_at":"2026-10-01 08:30:05"},{"id":"hot-demo-001","title":"AI Agent 正在进入企业工作流","category":"AI与科技","heat_score":40,"discussion_volume":6,"summary":"企业开始把 AI Agent 从单点聊天工具接入真实业务流程，热点讨论集中在可靠性、成本与自动化边界。","updated_at":"2026-10-21 10:47:28"},{"id":"f56154db-f68c-488f-bc54-0e9a50318cfa","title":"GPT-6 Sol (Max) Intelligence, Performance and Price Analysis","category":"AI与科技","heat_score":40,"discussion_volume":4,"summary":"综合各模型解读，可确认Artificial Analysis平台存在由theanonymousone于2026-09-22发布的《GPT-6 Sol (Max) 智能、性能与价格分析》报告，标题含(Max)。Kimi-K3还提到同日存在GPT-6 Luna及三份Claude Opus 5.5相关报告，但其他模型未提及。所有模型均表示无法获取具体基准得分、性能数据与定价信息，需查阅原始页面核实。各模型对核心事实的描述高度一致，未发现实质性分歧，仅信息详略不同。未确定项包括具体基准结果、性能指标、价格、对比结论及与其他模型的关系。","updated_at":"2026-09-30 05:33:22"}],"canonical_url":"http://www.aigc.guru/hot/topic/d4d274ff-74cf-47a8-a6c6-0fd5ee6450cb","discussion":{"comments":[],"next_cursor":null,"has_more":false}}}