[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"news-slug-liquid-ai-d1-ultra-decision-index":3,"topics-all":38,"news-related-e74f9194-9075-467a-95db-747a217fffa6":57},{"id":4,"title":5,"summary":6,"content":7,"original_url":8,"source_id":9,"tags":10,"translations":24,"news_slug":31,"published_at":32,"created_at":33,"modified_at":34,"is_published":35,"publish_type":36,"image_url":14,"view_count":37},"e74f9194-9075-467a-95db-747a217fffa6","d1-ultra登顶决策榜,但账单会乘法翻倍","Liquid AI闭源d1-ultra在Artificial Analysis决策指数拿下65分,官方称领先至少8分,输入每百万token收0.042美元。但零输出token不等于免费:图像多问题输入会让账单乘法翻倍,单次前向决策的过度自信也未解。","10月15日,Liquid AI上线闭源旗舰决策模型d1-ultra。这是d1系列的第二步:10月7日他们开源了端侧d1-3B与d1-omni-600M,八天后再把商用闭源版挂上Liquid API,并预告即将进入Vercel与OpenRouter。时机踩得很准——Jev式决策模型9月爆火之后,OpenAI、微软相继入场,Liquid AI这步棋把「不写长文、只给判定」的赛道又推高了一档。\n\n## 65分登顶,输入价0.042美元\n\n据[Liquid AI官方博客](https:\u002F\u002Fwww.liquid.ai\u002Fblog\u002Fd1-ultra),d1-ultra在Artificial Analysis新设立的Decision Index上拿到65分,领先其余所有决策模型至少8分;中文分析媒体ic.work援引的榜单数据显示,Jev 1.13为58分,微软同类模型54分。官方称在112项评估中d1-ultra全面超过上代d1与Jev 1.14,并专门针对校准做了训练——他们发现多数决策模型过度自信,而d1-ultra输出的概率更可信。定价按输入token计,每百万0.042美元,零输出token。作为参照,开源版d1-3B已在RTX 4090上做到8毫秒应答,闭源版则把决策质量再抬一层。\n\n## 省钱是卖点,计费是暗坑\n\n零输出token不等于免单。官方计费文档写明:图像按每32×32像素块1.5个token计数,一张1024×1024的图就是1536个token;更关键的是,「每个问题都作为独立prompt计费,包含其携带的全部图像」。[ic.work的分析](https:\u002F\u002Fwww.ic.work\u002Farticle\u002Fliquid-ai-releases-d1-decision-model-analysis)算了笔账:在一张图上挂3个质检问题,这张图的1536个token就会被足额计费3次。单次请求最多8张图、1万个patch——极低单价遇上问题数乘数,复合业务场景里账单可能瞬间反噬。\n\n## 开源版的成色与泛化雪崩\n\n开源一侧的成绩单更冷静。ic.work指出,d1-3B在11个公开图像基准平均74.1分,只比基座LFM2.5-VL-3B的73.9高0.2;587M的d1-omni-600M在Decision Index仅15.95分,模态互斥,基本是仓促放出的早期检查点。更值得警惕的是「一次前向定生死」的泛化风险:工程实测中,该架构原型域内准确率77.3%,撞上未见过的域外分布直接跌到54.6%,接近抛硬币;同类竞品在Banking77上的校准误差达0.098,面对无信息输入仍自信输出0.9以上。自回归模型写错了还能回头改,单次直出的模型错过就是错过。\n\n## 所以呢\n\nd1系列的定位其实很清晰:当智能体门口的毫秒级分诊员——工单该给哪个团队、内容有没有风险,一次前向出结果,长上下文推理留给生成式主模型,这个分工成立。但若被0.042美元的价签吸引,直接让单次决策模型接管工业闭环,请记住:它不生成token,也不生成第二次机会。开源背景见[Hugging Face博客](https:\u002F\u002Fhuggingface.co\u002Fblog\u002FLiquidAI\u002Fopen-d1)。","https:\u002F\u002Fwww.liquid.ai\u002Fblog\u002Fd1-ultra","511bb1e6-a31f-4dc1-929b-9a7582e67447",[11,15,18,21],{"id":12,"name":13,"slug":13,"description":14,"color":14},"fca9258a-9430-455a-b95d-b9fae5e373a8","ai-inference",null,{"id":16,"name":17,"slug":17,"description":14,"color":14},"120fa59a-ff6f-4537-9bf5-f818df636a0e","benchmark",{"id":19,"name":20,"slug":20,"description":14,"color":14},"7e89b5cc-57db-4f37-bc6d-28919a73931c","model-release",{"id":22,"name":23,"slug":23,"description":14,"color":14},"499f4b56-819d-49a3-9609-33e775143b86","multimodal",[25],{"id":26,"lang":27,"title":28,"summary":29,"content":30},"2998c3c3-feca-4253-81a5-db703f99c427","en","d1-ultra Tops Decision Index, But Watch the Bill","Liquid AI's d1-ultra scores 65 on the Decision Index, 8+ ahead, at $0.042\u002FM input tokens. Zero output is not free: image questions multiply bills.","On October 15, Liquid AI launched d1-ultra, its closed-source flagship decision model. It is the second move in the d1 playbook: on October 7 the company open-sourced the edge-sized d1-3B and d1-omni-600M, and eight days later put a commercial closed-source version on the Liquid API, with Vercel and OpenRouter integrations promised soon. The timing is deliberate — after Jev-style decision models went viral in September, OpenAI and Microsoft piled in, and Liquid AI has now pushed the \"no long-form text, just verdicts\" track up another notch.\n\n## 65 points on top, $0.042 per million tokens\n\nAccording to the [official Liquid AI blog](https:\u002F\u002Fwww.liquid.ai\u002Fblog\u002Fd1-ultra), d1-ultra scores 65 on the newly established Artificial Analysis Decision Index, leading every other decision model by at least 8 points; figures cited by Chinese outlet ic.work show Jev 1.13 at 58 and Microsoft's comparable model at 54. Liquid AI says d1-ultra beats its predecessor d1 and Jev 1.14 across 112 unique evaluations, and was specifically trained for calibration — the team found most decision models are overconfident, while d1-ultra produces more trustworthy probabilities. Pricing is input-only at $0.042 per million tokens, with zero output tokens. For reference, the open-weight d1-3B already answers in 8 ms on an RTX 4090; the closed-source version lifts decision quality another tier.\n\n## Cheap tokens, expensive questions\n\nZero output tokens does not mean free. The official pricing docs state that images count at 1.5 tokens per 32×32-pixel patch — a 1024×1024 image costs 1,536 tokens — and, crucially, \"each question is billed as its own prompt, including its text and all images.\" [ic.work's analysis](https:\u002F\u002Fwww.ic.work\u002Farticle\u002Fliquid-ai-releases-d1-decision-model-analysis) did the math: hang 3 quality-inspection questions on one image, and that image's 1,536 tokens get billed in full three times. A single request takes at most 8 images and 10,000 patches — an extremely low unit price multiplied by the question count can make bills rebound violently in composite business scenarios.\n\n## The open-weight reality and the generalization cliff\n\nThe open-source side tells a more sobering story. ic.work notes that d1-3B averages just 74.1 across 11 public image benchmarks, only 0.2 above its LFM2.5-VL-3B backbone; the 587M d1-omni-600M manages only 15.95 on the Decision Index, with mutually exclusive modalities — essentially an early checkpoint rushed out. More alarming is the generalization risk of the \"one forward pass, one verdict\" design: in engineering tests, a prototype of this architecture scored 77.3% in-domain but crashed to 54.6% on unseen out-of-domain distributions, close to a coin flip; comparable competitors reach a calibration error of 0.098 on Banking77 and still confidently output 0.9-plus on uninformative inputs. An autoregressive model that misspeaks can walk it back; a single-pass model that misses, misses.\n\n## So what\n\nThe d1 lineup's positioning is actually clear: work as the millisecond triage clerk at the agent's front door — which team gets this ticket, whether this content is risky — one forward pass, one answer, while long-context reasoning stays with the generative main model. That division of labor makes sense. But if the $0.042 price tag tempts you into letting a single-pass decision model take over an industrial control loop, remember: it generates no tokens, and it generates no second chances. Open-source background in the [Hugging Face blog](https:\u002F\u002Fhuggingface.co\u002Fblog\u002FLiquidAI\u002Fopen-d1).","liquid-ai-d1-ultra-decision-index","2026-10-20T13:10:00Z","2026-10-20T13:11:25.036857Z","2026-10-20T13:11:25.036866Z",true,"agent",73,[39,48],{"slug":40,"tag_slug":40,"title_zh":41,"title_en":42,"intro_zh":43,"intro_en":44,"id":45,"is_active":35,"created_at":46,"modified_at":47},"ai-for-science","AI for Science 2026：从 UniPert 到 GPT-Rosalind 的硬核进化","AI for Science 2026: from UniPert to GPT-Rosalind","生命科学、化学材料、物理世界模型——AI 正在从\"语言工具\"变成\"实验伙伴\"。本专题收录 AI 在三大科学方向的关键节点：UniPert 统一基因与化学扰动空间、GPT-Rosalind 端到端生命科学推理、达摩院 AI 智能体 28 小时找到 4 种超导新材料、Anthropic Claude Science 把工作台做成标准品。","From language tool to lab partner — AI is reshaping life sciences, chemistry\u002Fmaterials, and physical world models. This topic covers the key milestones: UniPert unifying genetic-chemical perturbation spaces, GPT-Rosalind's end-to-end life-sciences reasoning, DAMO's AI agent discovering 4 superconducting materials in 28 hours, and Anthropic's Claude Science workbench going mainstream.","988a4300-5fab-41c4-b5d8-63711a2dc757","2026-09-10T01:34:15.296649Z","2026-09-10T01:34:15.296663Z",{"slug":49,"tag_slug":49,"title_zh":50,"title_en":51,"intro_zh":52,"intro_en":53,"id":54,"is_active":35,"created_at":55,"modified_at":56},"h3-series","MiniMax H3 系列：从开源权重到 35 倍吞吐","MiniMax H3 Series: from open weights to 35x throughput","MiniMax H3 自 2026 年 8 月开源以来节奏密集：官方把生成、参考与编辑收回一个模型；ComfyUI 当天压进 RTX 3060；摩尔线程 3 小时完成国产 GPU 适配；fal 后训练版把吞吐拉到 35 倍；FastH3 蒸馏再砍推理成本。本专题持续追踪 H3 的发布—开源—蒸馏—部署全链路。","Since MiniMax open-sourced H3 in August 2026 the pace has been relentless: one unified omni-modal model, same-day ComfyUI support down to an RTX 3060, a 3-hour Day-0 port to Moore Threads GPUs, fal's post-trained H3 Max at 35x throughput, and FastH3 distillation cutting inference cost further. This topic tracks the full H3 chain — release, open weights, distillation, deployment.","83ef0daa-3c31-4cb3-86ed-e5ee58654d5f","2026-09-08T07:33:19.942193Z","2026-09-08T07:33:19.942209Z",{"items":58},[59,64,69,74,79,84],{"id":60,"title":61,"news_slug":62,"published_at":63},"d358cce6-d949-420c-9994-f266a55fb654","Granite Switch 4.1：聊天决策语音一端搞定","granite-switch-4-1-single-endpoint","2026-10-17T21:06:31+00:00",{"id":65,"title":66,"news_slug":67,"published_at":68},"df4159eb-60c1-4267-835b-0932d2a1ac72","d1-3B 开源：零token决策，8毫秒应答","liquid-ai-d1-open-edge-decision-models","2026-10-11T19:14:50+00:00",{"id":70,"title":71,"news_slug":72,"published_at":73},"240d8a4c-c025-4080-be14-06099f650a71","xAI 把视频价格打到底:1080p 蒸馏版 Lite 比 Veo 3.1 便宜 65%","grok-imagine-1-5-lite-cheaper-than-veo","2026-10-10T08:00:00+00:00",{"id":75,"title":76,"news_slug":77,"published_at":78},"6062d551-9068-4a9f-ae8e-4e99269cd838","Muse Voice Transcribe 发布:流式转写、20+ 说话人分离、端点检测,Meta 全塞进一个模型","meta-muse-voice-transcribe-streaming-asr","2026-09-05T13:11:00+00:00",{"id":80,"title":81,"news_slug":82,"published_at":83},"fdf05035-e02e-4954-8beb-c697ecc7975a","Cohere Parse 发布:$1.5 每千页的文档解析模型,ParseBench 79.2 超 Mistral OCR 4","cohere-parse-document-parsing-model","2026-08-28T13:30:00+00:00",{"id":85,"title":86,"news_slug":87,"published_at":88},"4c7f5330-3aff-458a-9ef5-f04cc5585703","微信视觉团队开源 WeMM 嵌入模型:2B 反超 8B 前基线,9B 达 MMEB-v2 80.6","wemm-embedding-wechat-multimodal","2026-08-26T21:07:30+00:00"]