Teachometry
Run运行

Results结果tutor-eval-v0.2a@0.2a.6

Results,
with the context intact.
结果,
连同语境一起保留。

Teachometry will publish benchmark results after the evaluation corpus, calibration, and validation processes are complete. We prioritize trustworthy, reproducible, and well-contextualized results over premature rankings.

Teachometry 会在评测语料、校准和验证流程完成后发布基准结果。我们优先保证结果可信、可复现且保留完整语境,而不是过早发布排行榜。

Read the methodology阅读方法论 Explore the benchmark探索基准

Evidence
over hype.
证据
胜过炒作。
A more transparent future
for AI tutoring.
让 AI 教学拥有
更透明的未来。

The back of an archive envelope on slate-blue bookcloth, its flap closed with ruby sealing wax pressed with the Teachometry seal. The printed form reads: dataset cohort tutor-eval-v0.2a@0.2a.6, benchmark version 0.1, calibrated public model runs none yet, status not yet published.
Cover封面 · Sealed until the evidence is ready.证据齐备之前,封存。

Public benchmark results公开基准结果

Not yet published.尚未发布。

No public model rankings are available in the current public artifact.当前公开产物尚未提供模型排名。 No calibrated public model runs yet.目前尚无校准后的公开模型运行记录。 Calibration and validation evidence remain incomplete.校准与验证证据仍不完整。

No ranking without evidence.
No score without context.
没有证据,就不做排名。
没有语境,就不报分数。

Future leaderboard schema未来排行榜结构

The table below shows the structure of the future public leaderboard.下表展示未来公开排行榜的结构。 It illustrates what information will be reported for each model. No model results are shown yet.它说明每个模型将公开哪些信息;目前尚未展示任何模型结果。

Future public model leaderboard schema未来公开模型排行榜结构
#Model模型Overall总体↑Correctness正确性Diagnosis诊断能力Guidance引导能力Adaptation适应能力Actionability可执行性Cost成本USD / 1KLatency延迟sBenchmark version基准版本Prompt versionPrompt 版本Dataset cohort数据集案例组
No public model rows are available yet.目前尚无公开模型记录。

No calibrated public model runs yet.目前尚无校准后的公开模型运行记录。 Synthetic demonstrations and preliminary model artifacts are not public rankings.合成演示和初步模型产物不构成公开排名。

Swipe horizontally to inspect the future schema. No result values are published in this preview.可横向滑动查看未来结构;此预览不发布任何结果值。

Readiness and evidence status准备度与证据状态

We follow a staged process to ensure that public results are meaningful and trustworthy.

我们采用分阶段流程,确保公开结果具有意义且值得信赖。

  1. 1
    Evaluation infrastructure评测基础设施

    Benchmark, evaluators, and scoring pipeline are available.基准、评估器和评分流程已经可用。

    Available可用
  2. 2
    Calibration infrastructure校准基础设施

    Tools and workflows for calibration are in place.校准所需工具与工作流已经就位。

    Available可用
  3. 3
    Community Review infrastructureCommunity Review 基础设施

    Public launch remains blocked; reviewer intake remains closed, and the real Community Review campaign has not started.公开发布仍被阻塞;reviewer intake 仍关闭,真实 Community Review 活动尚未启动。

    Private staging ready私有 staging 已就绪
  4. 4
    Human calibration (P5)人工校准(P5)

    No real human calibration data is available yet.目前尚无真实人工校准数据。

    Not started尚未开始
  5. 5
    Judge-vs-human validationJudge 与人工对照验证

    A Judge result is not presented as human-reference validation.Judge 结果不会被呈现为人工参考验证。

    Not completed尚未完成
  6. 6
    Statistical validation统计验证

    No statistical validation claim is made for the current benchmark.当前基准不主张已经完成统计验证。

    Not completed尚未完成

Example result view (preview)示例结果视图(预览)

This schema-only preview shows how individual model results will be presented.

这个仅展示结构的预览说明未来单个模型结果将如何呈现。

Model result contract模型结果契约

Future public model result未来公开模型结果

Versioned model identity and evidence fields; no public row is available yet.版本化模型身份与证据字段;目前尚无公开记录。

Overall总体
—
Correctness正确性
—
Diagnosis诊断能力
—
Guidance引导能力
—
Adaptation适应能力
—
Actionability可执行性
—

Subject breakdown学科拆分

  • History Or Social Studies历史或社会研究—
  • Language语言—
  • Mathematics数学—

Capability breakdown能力拆分

  • Answer Non Disclosure不泄露答案—
  • Check For Understanding检查理解—
  • Clear Next Step明确下一步—

Efficiency效率

  • Cost成本—
  • Latency延迟—
  • Tokenstoken 数—

Metadata & traceability元数据与可追踪性

  • Provider snapshot服务提供方快照—
  • Model version模型版本—
  • Benchmark version基准版本0.1
  • Prompt versionPrompt 版本—
  • Dataset cohort数据集案例组tutor-eval-v0.2a@0.2a.6
  • GenerationSpecIdGenerationSpecId—
  • Execution identity执行身份—

Generation identity and execution fields remain part of the future public contract.生成身份与执行字段仍是未来公开契约的一部分。

Our commitment我们的承诺

No ranking without evidence.
No score without context.
没有证据,就不做排名。
没有语境,就不报分数。

We are building an open, transparent, and community-driven benchmark for AI tutoring. Thank you for your interest and support.

我们正在构建一个开放、透明、由社区共同推动的 AI 教学基准。感谢你的关注与支持。