
mancer is an independent large language model inference service that routes your prompts to unfiltered models with no content filtering or behavioural guidelines. It offers a free tier for its standard model and paid tiers for faster or more capable ones, and exposes an OpenAI-compatible API that drops into tools like SillyTavern.
| 面向/指標 | ![]() | |
|---|---|---|
| 官方網域 | deepseek.com | mancer.tech |
| 定價模式 | 免費增值 | 免費增值 |
| 每月造訪次數 | 3.73億 (372,761,591) | 1708 (1,708) |
| 月流量變化量 | +1536.22萬 (+4.3%) | -7193 (-80.8%) |
| 主要市場與市佔 | 🇨🇳 中國 (50.5%) | 🇧🇷 巴西 (93.9%) |
| 每月直接造訪 | 2.92億 | 734 |
| 使用者評分 | 4.8 | 4.1 |
| 支援平台 | 網站 | 網站 |
由 DeepSeek-V4 與 R1 驅動的高效開源權重與託管推理 AI,token 成本極低。
依託稀疏混合專家架構(V4-Pro 為 1.6T 中的 49B 啟用參數),以極低延遲與記憶體開銷處理最多 100 萬 token。
由 DeepSeek-R1 開創並融入 V3.2 與 V4 思考模式,對複雜數學、邏輯與程式碼除錯展示逐步自驗證過程。
快取命中時,輸入 token 成本降至 Flash 的每百萬 0.003 美元與 V4-Pro 的每百萬 0.022 美元,離峰 UTC 時段費率再低 50%。
在 chat.deepseek.com 及 iOS/Android 上免費提供文件上傳、推理模式與網頁搜尋,核心模型無須付費牆。

mancer is an independent large language model inference service that routes your prompts to unfiltered models with no content filtering or behavioural guidelines. It offers a free tier for its standard model and paid tiers for faster or more capable ones, and exposes an OpenAI-compatible API that drops into tools like SillyTavern.
No content filtering and no behavioural guidelines are applied on the provider side. mancer runs the models and returns whatever they generate, leaving the choice of what to prompt to the user.
Access a capable model at no cost for everyday use, and move to paid models when you need faster response times, larger parameter sizes or stronger general ability.
An OpenAI-compatible text-generation API means chat front-ends such as SillyTavern and other OpenAI-style clients can point straight at mancer without bespoke integration work.
Sign-up, model selection and usage are a short, three-step flow designed so you can get a key and start prompting quickly rather than working through a heavy onboarding process.

提供巨大的上下文視窗與深度的 Google 生態系整合,但對於純推理任務感覺較不夠專精。
提供巨大的上下文視窗與深度的 Google 生態系整合,但對於純推理任務感覺較不夠專精。

免費 / 開源阿里雲第四代基礎大模型家族與全模態 AI 助手 — 涵蓋 Qwen 4 (Qwen4)、Qwen4-Max、Qwen4-Plus、Qwen4-Flash、Qwen4-27B 與 Qwen 3.8-Flash-Next 架構
免費 / 開源阿里雲第四代基礎大模型家族與全模態 AI 助手 — 涵蓋 Qwen 4 (Qwen4)、Qwen4-Max、Qwen4-Plus、Qwen4-Flash、Qwen4-27B 與 Qwen 3.8-Flash-Next 架構
免費 / 開源Enables running open-source models locally on consumer hardware, but lacks DeepSeek's cloud-scale hosted reasoning power.
免費 / 開源Enables running open-source models locally on consumer hardware, but lacks DeepSeek's cloud-scale hosted reasoning power.

智譜 AI 的 GLM 大模型開放平台,多模態 API 與代理開發一次到位