한국어EN
Evaluation & BenchmarksTerminal-Bench team

Terminal-Bench team releases Terminal-Bench 2.0 and Harbor evaluation harness

Made by AIWritten by AI (Claude) from the source below. Check anything important against the source.How it's made

Released Terminal-Bench 2.0 and Harbor, an evaluation harness

Why it matters

Evaluation infrastructure that adds cloud container execution and an RL/SFT rollout interface to agent CLI evaluation

Source — Primary sources first; secondary reports are labelled “Press”

Officialtbench.aihttps://www.tbench.ai/news/announcement-2-0

Added in Week 35, 2026 · See that week

On the map — ⑯ Evaluation line

  1. Previous stopGDPval
  2. This recordTerminal-Bench team
  3. Next stopTime Horizon 1.1
See it on the map

Same week

Google Cloud adds Auto Vector Embeddings and Auto Vector Index to AlloyDB AIOfficialGoogle Cloud announces Ironwood (TPU v7)OfficialMoonshot AI releases Kimi K2 ThinkingOfficialAnthropic publishes Code Execution with MCPOfficial

Nearby on the same axis

Andon Labs releases Vending-Bench 2 and the multi-agent Vending-Bench ArenaOfficialOpenAI releases GDPvalOfficialMeta releases GAIA2 and the open-source execution framework AREOfficialScale AI releases SWE-Bench ProPaper