Terminal-Bench Data
Agents solve terminal tasks in machine learning, data analysis, and scientific computing. Task verifiers grade the outputs.
Turing created 300 original tasks and evaluated three models.
- Made by
- Turing
- Released
- July 2026
- Kind
- Terminal tasks
- Level
- Expert
- Scored
- Checked automatically
- Answers
- Agent task
- Tasks
- Kept private
- Measures
- Capability
- Interaction
- Agentic
- Environment
- Terminal
- Items from
- Written by experts
- Refresh
- One fixed set
- Run by
- A host
- In views
- Agents
- Strongest model tested
- GPT-5.5 #21 on the capability index
- Last model added
- May 2026
- Scores refreshed
- November 20th, 2026
GPT-5.5 leads with 48.6% among models released by May 2026
Best score per model from the benchmark's own leaderboard, read on November 20th, 2026. Where a model ran with several settings or harnesses, we keep its best.
The newest model with a score here came out in May 2026, and nobody has tested a newer one. Among the 20 highest-ranked models on the capability index, these have no score yet:
- GPT-6.2 Astra
- Claude Fable 5.2
-
Gemini 4 Carbon
- GPT-6 Astra
- and 16 more
- 1 GPT-5.5 48.6%
- 2 Claude Opus 4.8 41.3%
- 3
Gemini 3.5 Flash 38.3%
Source: Official Terminal-Bench Data leaderboard
| # | Model | Score ± | Setting | Date | Source |
|---|---|---|---|---|---|
| 1 | GPT-5.5 | 48.6% ±1.4 | Terminus-2 | Apr 2026 | Official leaderboard |
| 2 | Claude Opus 4.8 | 41.3% ±1.4 | Terminus-2 | May 2026 | Official leaderboard |
| 3 | | 38.3% ±1.7 | Terminus-2 | May 2026 | Official leaderboard |
Rankings by part
We read 3 parts from the benchmark’s own leaderboard: Scientific Computing, Machine Learning and Data Analysis. Each ranks models on one part of the test. GPT-5.5 leads every one of them. Arrows show how far a model moved from its overall rank among the models we track.
Scientific Computing
3 models we track have a score here.
- 1 GPT-5.5 51.1%
- 2 Claude Opus 4.8 38.4%
- 3
Gemini 3.5 Flash 34.1%
Machine Learning
3 models we track have a score here.
- 1 GPT-5.5 44.1%
- 2 Claude Opus 4.8 42.3%
- 3
Gemini 3.5 Flash 39.9%
Data Analysis
3 models we track have a score here.
- 1 GPT-5.5 48.1%
- 2 Claude Opus 4.8 45.2%
- 3
Gemini 3.5 Flash 43.7%
Number one over time
Lab claims count from the day the card came out, independent runs from the model’s release.
Source: Official Terminal-Bench Data leaderboard
Scores over time
All the scores we have, one dot each, so a model tested at several settings appears more than once.
Source: Official Terminal-Bench Data leaderboard
No lab model card reports it yet
We read the headline results of 540 launch posts from 57 labs, and none of them list this benchmark.
More in Terminal agents
- SkillsBench In use 86-87 containerized expert tasks in 11 domains, run with and without Agent Skills. Verifier tests give pass rate and skill lift. GPT-5.5 66.5%
- Terminal-Bench Hard In use Agents complete 44 hard tasks from Terminal-Bench Core in the Terminus 2 harness. Artificial Analysis scores pass@1 averaged over 3 repeats.
Gemma 4 31B IT 36%
- Terminal-Bench 4.1 In use The Terminal-Bench 4.0 tasks with hardened verifiers: agents complete 66 hard tasks in containerized terminals across software, ML, science, security, and operations. Test scripts verify each result, scored as resolution rate. Copilot Fusion 71.2%
- Harbor-Index In use Agents solve about 80 hard tasks drawn from 29 benchmarks run through Harbor adapters, such as HLE, GSO and SWE-bench Pro. Each task keeps its source verifier. GPT-5.5 28.1%
- Terminal-Bench 2.1 Extended Not in model cards Agents complete long command-line tasks in containers, scored by verification tests. Mercor adds 99 new expert-written tasks to the 89 public ones and reports both scores.
Muse Wonder 54%
- Terminal-Bench Pro Not in model cards Agents solve 400 terminal tasks across eight domains. Harbor distributes the 200 public tasks. Alibaba keeps the other 200 for private evaluation. Claude Opus 4.5 51%
Data audit
8 of 9 checks pass, 1 warning. Look first at the Newest score check. The newest came 6 months ago, in May 2026, for Claude Opus 4.8. Every benchmark
Catalog record 2 pass
- Pass Task access The tasks are private.
- Pass Confidence Checked against a primary source.
Scores 1 warning 2 pass 2 notes
- Pass Independent runs Scores from the official leaderboard (3 entries).
- Note Models scored Only 3 models we track.
- Warning Newest score The newest came 6 months ago, in May 2026, for Claude Opus 4.8.
- Note Frontier models None of the top 10 has a score. The strongest tested is GPT-5.5 (#21).
- Pass Headroom The best is 48.6%, under 85%, so it still separates models.
Leaderboard import 3 pass
- Pass Importer We read the board on labs.turing.com: 3 rows for models we track at the last read.
- Pass Last read November 20th, 2026.
- Pass Read worked The last read worked.
Consistency 1 pass 1 note
- Note Lab claims No model has both a lab number and an independent run.
- Pass Known issues No published audit found problems.