跳到主要内容Skip to main content
投机取巧 Speculative Speculation

Kwaipilot（快手）Kwaipilot (Kuaishou)　·　当前代current generation KAT-Coder-V2.5KAT-Coder-V2.5（2026-07）

KAT CoderKAT Coder模型家族谱系卡model family lineage card

快手 Kwaipilot 的代码家族：API 版闭源、Dev 版开放权重。它是本站收录的「派生模型」代表——官方自己声明基座是 Qwen3.6-35B-A3B，选它等于选 Qwen 谱系 + 快手代码后训练。Kuaishou Kwaipilot's coding line: the API build stays closed, the Dev build is open-weight. It is this site's representative "derived model" — the vendor itself declares Qwen3.6-35B-A3B as its base, so choosing it means choosing the Qwen lineage plus Kuaishou's code post-training.

本站已收录该家族的量化对比：This site covers this family's quantisations: KAT Coder V2.5KAT Coder V2.5

先说清楚：这个站的数据怎么来的First, where this site's data comes from

✅ 实测字段Measured fields

直接从 Hugging Face 公开 API 读取的客观数据：文件体积（字节）、上下文长度、架构、许可、下载量、更新时间。Objective data read straight from the Hugging Face public API: file sizes in bytes, context length, architecture, licence, downloads, last updated.

✅ 推导Derived

由实测字段按公开规则算出：每权重比特数（依 llama.cpp 命名标准）、档位归属、显存需求区间。推导规则在Computed from the measured fields by published rules: bits per weight (per the llama.cpp naming convention), tier assignment, VRAM requirement range. The rules are fully published in METHODOLOGY.md 完整公开，任何人可复算。 — anyone can recompute them.

❌ 我们不说的What we do not claim

哪个量化质量更高、哪个更快、哪个模型更强——这些需要实测，本站不跑 benchmark，所以不装作知道。凡我们给出的主观判断，一律显式标注「我们的口径」。Which quantisation is higher quality, which is faster, which model is stronger — these need measurement. This site does not run benchmarks, so it does not pretend to know. Every subjective judgement we do offer is explicitly marked as our own reading.

本站This site does 不下载、不托管任何模型权重文件not download or host any model weight files。所有数字都来自公开元数据，点击即可回到发布页核对。. Every figure comes from public metadata, and clicking it takes you back to the release page to verify. 采集时间Collected: 2026-10-30（数据快照，非实时） (a snapshot, not live)

中文生态China availability 国产 / 国内可直接用 / 备案状态 · 快照domestic / direct access / CAC filing · snapshot 2026-10-20

  • 出品属地Origin　国产domestic　出品方 Kwaipilot 是快手（北京）的 AI 团队——官方 HF 组织页自述「the AI explorers from Kuaishou」。Kwaipilot is the AI team of Kuaishou (Beijing) — its official HF org page says "the AI explorers from Kuaishou".
  • 国内下载Direct access　国内可直连direct OK　ModelScope 官方镜像直连可达（2026-10-20 实测 200）；Hugging Face 直连不可达。Official ModelScope mirror reachable (measured 200); Hugging Face direct unreachable.　·　国内镜像入口domestic mirror
  • 备案状态CAC filing　未见于备案名单not in filing list：「KAT Coder」未出现在网信办公示名单（2026-10-20 检索）；快手旗下「快意大模型」「可图大模型」「可灵」等已备案，但本站不替 KAT Coder 推断。“KAT Coder” does not appear in the CAC public registry (searched 2026-10-20); Kuaishou’s “Kuaiyi”, “Ketu” and “Kling” models are filed, but this site does not infer a filing for KAT Coder.

备案信息以网信办《生成式人工智能服务已备案信息》公示名单为准（按批次滚动更新，本站只抄编号不解读）；「国内可直连」是本站从大陆网络直连实测的结果，不是厂商承诺。Filing status is quoted from the CAC public registry (updated batch by batch; this site copies the number and does not interpret it). "Direct access" is this site's own reachability measurement from a mainland network, not a vendor promise.

代次Generations

  • KAT-Coder-V2.5KAT-Coder-V2.52026-07当前代current

    当前代。官方 README：V2.5 于 2026-07 发布（闭源 API 版），开放权重的 V2.5-Dev 为 35B 总参 / 3B 激活 MoE，经 SFT/RL 训练；模型卡声明派生自 Qwen3.6-35B-A3B。Current generation. Official README: V2.5 shipped July 2026 (closed API); the open-weight V2.5-Dev is a 35B-total / 3B-active MoE trained with SFT/RL; the model card declares it derived from Qwen3.6-35B-A3B.

    官方页面 ↗official page ↗
  • KAT-V1KAT-V12025

    更早一代（KAT-V1-40B 仓 2025-08 出现在官方 org）。Earlier line (the KAT-V1-40B repo appeared in the official org in August 2025).

    官方页面 ↗official page ↗

派生标记Derived from 按官方模型卡 base_model 声明per the official model-card base_model declaration

官方模型卡 cardData 声明 base_model: Qwen3.6-35B-A3B —— 本站按官方 base_model 口径把它标为「派生自 Qwen 系」：权重谱系上它是 Qwen3.6-35B-A3B 的 SFT/RL 后训练产物，不是从零预训练。The official model card declares base_model: Qwen3.6-35B-A3B — under this site's official base_model convention it is marked as derived from the Qwen line: on the weight lineage it is an SFT/RL post-train of Qwen3.6-35B-A3B, not trained from scratch.　基座base：Qwen3.6-35B-A3B　出处 ↗source ↗

尺寸档Size tiers

  • 代码专用 35B-A3B（35B 总参 / 3B 激活）Code-focused 35B-A3B (35B total / 3B active)

许可Licences

官方能力声明（带出处）Official capability claims (with sources)

  • 官方 README：「an MOE model with a total parameter count of 35B and 3B activated parameters」「Through SFT/RL training, KAT-Coder-V2.5-Dev achieves SOTA results … among models with similar parameter scales」——SOTA 为厂商自评基准口径。Official README: "an MOE model with a total parameter count of 35B and 3B activated parameters"; "Through SFT/RL training, KAT-Coder-V2.5-Dev achieves SOTA results … among models with similar parameter scales" — the SOTA claim is vendor-run benchmark wording.出处 ↗source ↗

以上均为厂商官方页面原文或官方自评基准的转述；涉及横评结论的，本站不背书——本站不跑 benchmark。The above are quotations of official pages or transcriptions of vendor-run benchmarks; this site does not endorse cross-model conclusions — it runs no benchmarks.

出处Sources

核验快照：2026-10-30 · 本站不实测、不排名，未知就写未知。Verification snapshot: 2026-10-30 · no benchmarks, no rankings — unknown stays unknown.