我们的平台The platform

专属模型全生命周期管理Lifecycle management for dedicated models

数据、评分、训练、报告、上线和迭代,都在同一个控制台里,团队和智能体用同一套规则协作。Data, grading, training, reports, serving and improvement in one console, where teams and agents follow the same rules.

Super Model console overview: models in use, training runs, reports that reached parity, monthly savings, and the AI summary of the latest report

控制台界面The console

上线后与通用大模型统一路由Routed together with general models

路由层按质量、成本和延迟为每次请求挑模型。专属模型拿不准的题,自动回退给通用模型。The router picks a model for every request by quality, cost and latency, and falls back to a general model when the dedicated model is unsure.

业务应用Business app model: ent-copy Super Router 质量、成本、延迟quality, cost, latency 专属模型 v1Dedicated model v190% 专属模型 v2Dedicated model v2灰度 10%canary 10% 候选 v3Candidate v3影子,只打分shadow, scored only 通用大模型General models难题回退fallback on hard cases
$ factory train train-5ed17a9d --purpose train:content-copy \
    --base Qwen/Qwen2.5-7B-Instruct --method grpo
✓ 预检通过preflight passed  grant, no_leak, graders, licence, budget
$ 最坏花费worst case $49.90, 上限cap $100
● 2×H100  step 30/30  test 0.923
✓ 已登记模型版本model version registered  mv-grpo-20261110

$ factory report build eval-outputs --model-version mv-grpo-20261110
✓ rpt-20261110-f9cf13 已冻结,达到对等门槛frozen, parity reached

四种接入方式,同一套权限与规则Four entry points, one set of rules

企业的智能体可以自行建数据集、发起训练、读报告;它想上线时只会提交审批,由人来拍板。Agents can build datasets, start training and read reports. When one wants to go live, it files an approval and a person decides.

控制台Console业务、专家和管理员For business, experts and admins
APIOpenAI 兼容的推理与控制接口OpenAI-compatible inference and control
命令行CLI工程师和自动化流水线For engineers and pipelines
MCP在智能体里直接驱动工厂Drive the factory from agents

支持主流开源底座Built on the leading open models

底座按许可证管理,每个底座带好自己的对话模板,训练和推理用同一套。Bases are cleared by licence, and each carries its own chat template, used the same way in training and serving.

Qwen 2.5 / 3Apache-2.0DeepSeek-R1-DistillMITGLM-4MITLlama 3.1 / 3.3CommunityMistralApache-2.0Gemma 3Gemma TermsPhi-4MITgpt-ossApache-2.0

关键操作需人工审批Key actions require approval

上线审批Go-live approval

灰度、影子流量和全量都由管理员批准,申请人和批准人都记录在案。Canary, shadow traffic and full rollout are approved by an admin, with both names on record.

花费有上限Spending has a cap

开卡前先算最坏花费,超出预算不启动;机器到时限自动释放。Worst-case cost is checked before any GPU starts, and every machine releases itself at its time limit.

全程可追溯Everything is traceable

谁在什么时候用哪份数据做了什么,逐条可查,随时一键回滚。Who used which data for what, and when, line by line, with one-action rollback.

建议从一个高频任务开始评估Start the evaluation with one frequent task

评估从数据体检开始:判断该任务能否训练、是否值得训练,并给出预期成本与回本周期。The evaluation starts with a data check-up: whether the task can be trained, whether it is worth it, and the expected cost and payback.