Skip to main content

Benchmark profile

PostTrain Bench

A software-engineering benchmark for post-training infrastructure and implementation tasks, evaluated through the official Harbor implementation.

Data verified

Benchmark score on PostTrain Bench — July 16, 2026

BenchLM mirrors the published score view for PostTrain Bench. Kimi K3 leads the public snapshot at 36.6%. BenchLM does not use these results to rank models overall.

1 modelCodingCurrentDisplay onlyUpdated July 16, 2026

Benchmark score table (1 model)

Score
1
Kimi K3Moonshot AI · Closed
36.6%

About PostTrain Bench

Year

2026

Tasks

Post-training software-engineering tasks

Format

Harbor agent evaluation

Difficulty

Frontier software engineering

Moonshot evaluates Kimi K3 at maximum reasoning effort with the official Harbor implementation and Claude Code harness, averaged over three runs. BenchLM stores the provider-exact value as display-only launch evidence.

BenchLM freshness & provenance

Version

PostTrain Bench 2026

Refresh cadence

Quarterly

Staleness state

Current

Question availability

Public benchmark set

CurrentDisplay only

BenchLM uses freshness metadata to decide whether a benchmark should still be treated as a strong differentiator, a benchmark to watch, or a display-only reference. For the full scoring policy, see the BenchLM methodology page.

FAQ

What does PostTrain Bench measure?

A software-engineering benchmark for post-training infrastructure and implementation tasks, evaluated through the official Harbor implementation.

Which model scores highest on PostTrain Bench?

Kimi K3 by Moonshot AI currently leads with a score of 36.6% on PostTrain Bench.

How many models are evaluated on PostTrain Bench?

1 AI models have been evaluated on PostTrain Bench on BenchLM.

Last updated: July 16, 2026 · BenchLM version PostTrain Bench 2026

The AI models change fast. We track them for you.

A weekly brief for engineers and researchers covering new models, ranking shifts, and pricing changes.

Free. No spam. Unsubscribe anytime.