Codesota · Agentic AI · Autonomous Coding · Terminal-Bench 2.0Tasks/Agentic AI/Autonomous Coding
Autonomous Coding · benchmark dataset · 2026 · TERMINAL

Terminal-Bench 2.0.

Terminal-agent benchmark for software engineering, machine learning, security, data science, system administration, file operations, and related terminal workflows. Scores measure the agent harness and underlying model as one system.

Paper ↗Download datasetSubmit a result ↵
§ 01 · Leaderboard

Best published scores.

20 results indexed across 1 metric. Shaded row marks current SOTA; ties broken by submission date.

Primary
accuracy · higher is better
accuracy· primary
20 rows
#ModelOrgSubmittedPaper / codeaccuracy
01Codex / GPT-5.5APIOpenAIApr 2026terminal-bench-official82
02ForgeCode / GPT-5.4APIForgeCodeApr 2026terminal-bench-official81.80
03TongAgents / Gemini 3.1 ProAPITongAgentsApr 2026terminal-bench-official80.20
04ForgeCode / Claude Opus 4.6APIForgeCodeApr 2026terminal-bench-official79.80
05SageAgent / GPT-5.3-CodexAPISageAgentApr 2026terminal-bench-official78.40
06ForgeCode / Gemini 3.1 ProAPIForgeCodeApr 2026terminal-bench-official78.40
07Droid / GPT-5.3-CodexAPIDroidApr 2026terminal-bench-official77.30
08Capy / Claude Opus 4.6APICapyApr 2026terminal-bench-official75.30
09Simple Codex / GPT-5.3-CodexAPIOpenAIApr 2026terminal-bench-official75.10
10Terminus-KIRA / Gemini 3.1 ProAPITerminus-KIRAApr 2026terminal-bench-official74.80
11Terminus-KIRA / Claude Opus 4.6APITerminus-KIRAApr 2026terminal-bench-official74.70
12Mux / GPT-5.3-CodexAPIMuxApr 2026terminal-bench-official74.60
13MAYA-V2 / Claude 4.6 OpusAPIMAYAApr 2026terminal-bench-official72.10
14TongAgents / Claude Opus 4.6APITongAgentsApr 2026terminal-bench-official71.90
15Junie CLI / MultipleAPIJetBrainsApr 2026terminal-bench-official71
16CodeBrain-1 / GPT-5.3-CodexAPICodeBrainApr 2026terminal-bench-official70.30
17Droid / Claude Opus 4.6APIDroidApr 2026terminal-bench-official69.90
18Ante / Gemini 3 ProAPIAnteApr 2026terminal-bench-official69.40
19IndusAGI Coding Agent / GPT-5.3-CodexAPIIndusAGIApr 2026terminal-bench-official69.10
20Crux / Claude Opus 4.6APICruxApr 2026terminal-bench-official66.90
Fig 2 · Rows sorted by score within each metric. Shaded row marks SOTA. Dates reflect model or paper release where available, otherwise the date Codesota accessed the source.
§ 03 · Progress

1 steps
of state of the art.

Each row below marks a model that broke the previous record on accuracy. Intermediate submissions are kept in the leaderboard above; only SOTA-setting entries are re-listed here.

Higher scores win. Each subsequent entry improved upon the previous best.

SOTA line · accuracy
  1. Apr 27, 2026Codex / GPT-5.5OpenAI82
Fig 3 · SOTA-setting models only. 1 entries span Apr 2026 → Apr 2026.
§ 06 · Contribute

Have a score that beats
this table?

Submit a checkpoint and a reproduction script. We will run it, publish the score, and — if it takes the top — annotate the step on the progress chart with your name.

Submit a result ↵Read submission guide
What a submission needs
  • 01A public checkpoint or API endpoint
  • 02A reproduction script with frozen commit + seed
  • 03Declared evaluation environment (Python, deps)
  • 04One row per metric declared by this dataset
  • 05A contact so we can follow up on discrepancies