{"slug":"terminal-bench-2-1-benchmark-leaderboard","name":"Terminal-Bench 2.1 Benchmark Leaderboard","summary":"Artificial Analysis independent leaderboard for Terminal-Bench 2.1, a verified refresh of the Terminal-Bench 2.0 agentic terminal benchmark. It covers the same 89 curated tasks in software engineering, system administration, data processing, model training, and security, with env","publisher":"Artificial Analysis","homepage":"https://artificialanalysis.ai/evaluations/terminalbench-2-1","kind":"eval","hasApi":false,"hasMcp":false,"costRange":"unknown","costNotes":"Unknown: No price is stated for viewing or using this leaderboard. The page reports average evaluation cost per task in USD, split by input, cache hit, cache write, reasoning, and answer token pricing where canonical token counts are available, but that is a reported model-run cost, not a product price.","functions":["cache","index","reason","test","compose","evaluate","generate","policy"],"slices":[{"functionId":"evaluate","label":"Independent Terminal-Bench 2.1 scoring","note":"Scores agents on 89 curated terminal tasks using programmatic verification, the Terminus 2 harness, and an e2b sandbox, reported as pass@1 averaged over 3 repeats per task.","cost":"Unknown"},{"functionId":"cache","label":"cache","note":"Inferred from executor","cost":"Unknown"},{"functionId":"index","label":"index","note":"Inferred from executor","cost":"Unknown"},{"functionId":"reason","label":"reason","note":"Inferred from executor","cost":"Unknown"},{"functionId":"test","label":"test","note":"Inferred from executor","cost":"Unknown"},{"functionId":"compose","label":"compose","note":"Inferred from executor","cost":"Unknown"},{"functionId":"generate","label":"generate","note":"Inferred from executor","cost":"Unknown"},{"functionId":"policy","label":"policy","note":"Inferred from executor","cost":"Unknown"}],"interfaces":[],"composabilityRationale":"The retrieved page is a public evaluation leaderboard with methodology, score, cost, token, and speed views. It does not declare an API, MCP server, export interface, or other programmatic access, so composability beyond reading the page is unknown.","signals":[],"functionProbabilities":[{"functionId":"cache","value":0.8,"source":"https://artificialanalysis.ai/evaluations/terminalbench-2-1","method":"lexical","observedAt":"2026-10-19T01:25:18.728Z"},{"functionId":"index","value":0.8,"source":"https://artificialanalysis.ai/evaluations/terminalbench-2-1","method":"lexical","observedAt":"2026-10-19T01:25:18.728Z"},{"functionId":"reason","value":0.8,"source":"https://artificialanalysis.ai/evaluations/terminalbench-2-1","method":"lexical","observedAt":"2026-10-19T01:25:18.728Z"},{"functionId":"test","value":0.8,"source":"https://artificialanalysis.ai/evaluations/terminalbench-2-1","method":"lexical","observedAt":"2026-10-19T01:25:18.728Z"},{"functionId":"compose","value":0.72,"source":"https://artificialanalysis.ai/evaluations/terminalbench-2-1","method":"lexical","observedAt":"2026-10-19T01:25:18.728Z"},{"functionId":"evaluate","value":0.72,"source":"https://artificialanalysis.ai/evaluations/terminalbench-2-1","method":"lexical","observedAt":"2026-10-19T01:25:18.728Z"},{"functionId":"generate","value":0.72,"source":"https://artificialanalysis.ai/evaluations/terminalbench-2-1","method":"lexical","observedAt":"2026-10-19T01:25:18.728Z"},{"functionId":"policy","value":0.72,"source":"https://artificialanalysis.ai/evaluations/terminalbench-2-1","method":"lexical","observedAt":"2026-10-19T01:25:18.728Z"}],"claimVerification":[{"key":"mcp","status":"unverified","asserted":"false","evidenceSource":"https://artificialanalysis.ai/evaluations/terminalbench-2-1","method":"fallback"},{"key":"api","status":"unverified","asserted":"false","evidenceSource":"https://artificialanalysis.ai/evaluations/terminalbench-2-1","method":"fallback"},{"key":"cost","status":"unverified","asserted":"unknown","evidenceSource":"https://artificialanalysis.ai/evaluations/terminalbench-2-1","method":"fallback"},{"key":"license","status":"unverified","asserted":"","evidenceSource":"https://artificialanalysis.ai/evaluations/terminalbench-2-1","method":"fallback"}],"tweets":[],"notes":[],"origin":"intake","createdAt":"2026-10-19T01:25:19.515Z","updatedAt":"2026-10-19T01:25:19.515Z","sources":[{"kind":"web","label":"executor","url":"https://artificialanalysis.ai/evaluations/terminalbench-2-1","excerpt":"Artificial Analysis\n\nAll evaluations\n\n# Terminal-Bench 2.1 Benchmark Leaderboard\n\nA verified refresh of Terminal-Bench 2.0 — 89 curated tasks across software engineering, system administration, data processing, model training, and security,","at":"2026-10-19T01:25:19.515Z"}],"recordId":"cmp_terminal-bench-2-1-benchmark-leaderboard","claims":{"name":"Terminal-Bench 2.1 Benchmark Leaderboard","summary":"Artificial Analysis independent leaderboard for Terminal-Bench 2.1, a verified refresh of the Terminal-Bench 2.0 agentic terminal benchmark. It covers the same 89 curated tasks in software engineering, system administration, data processing, model training, and security, with env","kind":"eval","publisher":"Artificial Analysis","homepage":"https://artificialanalysis.ai/evaluations/terminalbench-2-1","costRange":"unknown","costNotes":"Unknown: No price is stated for viewing or using this leaderboard. The page reports average evaluation cost per task in USD, split by input, cache hit, cache write, reasoning, and answer token pricing where canonical token counts are available, but that is a reported model-run cost, not a product price.","functions":["cache","index","reason","test","compose","evaluate","generate","policy"],"slices":[{"functionId":"evaluate","label":"Independent Terminal-Bench 2.1 scoring","note":"Scores agents on 89 curated terminal tasks using programmatic verification, the Terminus 2 harness, and an e2b sandbox, reported as pass@1 averaged over 3 repeats per task.","cost":"Unknown"},{"functionId":"cache","label":"cache","note":"Inferred from executor","cost":"Unknown"},{"functionId":"index","label":"index","note":"Inferred from executor","cost":"Unknown"},{"functionId":"reason","label":"reason","note":"Inferred from executor","cost":"Unknown"},{"functionId":"test","label":"test","note":"Inferred from executor","cost":"Unknown"},{"functionId":"compose","label":"compose","note":"Inferred from executor","cost":"Unknown"},{"functionId":"generate","label":"generate","note":"Inferred from executor","cost":"Unknown"},{"functionId":"policy","label":"policy","note":"Inferred from executor","cost":"Unknown"}],"interfaces":[],"composabilityRationale":"The retrieved page is a public evaluation leaderboard with methodology, score, cost, token, and speed views. It does not declare an API, MCP server, export interface, or other programmatic access, so composability beyond reading the page is unknown."},"relations":[{"from":"terminal-bench-2-1-benchmark-leaderboard","to":"fix-cold-start-latency-in-ai-sandboxes-techniques","kind":"fills-same-slot-as","note":"Shared functions cache, compose, evaluate, reason, test. Can fill the same function slot."},{"from":"terminal-bench-2-1-benchmark-leaderboard","to":"chatgpt-scheduled-tasks","kind":"fills-same-slot-as","note":"Shared functions compose, evaluate, reason, test. Can fill the same function slot."},{"from":"terminal-bench-2-1-benchmark-leaderboard","to":"lm-evaluation-harness-pt","kind":"fills-same-slot-as","note":"Shared functions evaluate, generate, reason, test. Can fill the same function slot."},{"from":"terminal-bench-2-1-benchmark-leaderboard","to":"ai-agent-sandbox-comparison-2026-seven-platforms","kind":"fills-same-slot-as","note":"Shared functions compose, evaluate, policy, test. Can fill the same function slot."},{"from":"terminal-bench-2-1-benchmark-leaderboard","to":"postgresql","kind":"fills-same-slot-as","note":"Shared functions policy, test. Can fill the same function slot."},{"from":"terminal-bench-2-1-benchmark-leaderboard","to":"harness-arena","kind":"fills-same-slot-as","note":"Shared functions evaluate, generate, test. Can fill the same function slot."},{"from":"terminal-bench-2-1-benchmark-leaderboard","to":"webvoyager-leaderboard","kind":"fills-same-slot-as","note":"Shared functions cache, evaluate, policy. Can fill the same function slot."},{"from":"terminal-bench-2-1-benchmark-leaderboard","to":"the-agent","kind":"fills-same-slot-as","note":"Shared functions cache, evaluate, generate. Can fill the same function slot."}],"explain":{"score":3,"formula":"100 × sum(weight × value) over observed factors only. Unobserved factors are omitted, not zeroed.","factors":[{"id":"github_health","label":"GitHub health","weight":0.18,"why":"insufficient evidence","sources":[],"evidence":"insufficient"},{"id":"agent_use","label":"Agent use","weight":0.16,"why":"insufficient evidence","sources":[],"evidence":"insufficient"},{"id":"human_notes","label":"Human notes","weight":0.14,"why":"insufficient evidence","sources":[],"evidence":"insufficient"},{"id":"composability","label":"Composability","weight":0.16,"why":"insufficient evidence","sources":[],"evidence":"insufficient"},{"id":"checklist","label":"Checklist","weight":0.1,"value":0,"contribution":0,"why":"No MCP surface is declared. No API is declared. Cost is unknown. 0 of 3 checklist fields are complete.","sources":[],"evidence":"observed"},{"id":"evidence_quality","label":"Evidence quality","weight":0.14,"value":0.18107144632819594,"contribution":0.025350002485947435,"why":"1 sources cover 1 evidence types. The latest dated evidence is 2026-10-19. Freshness is omitted unless observed as a signal.","sources":[{"kind":"web","label":"executor","url":"https://artificialanalysis.ai/evaluations/terminalbench-2-1","excerpt":"Artificial Analysis\n\nAll evaluations\n\n# Terminal-Bench 2.1 Benchmark Leaderboard\n\nA verified refresh of Terminal-Bench 2.0 — 89 curated tasks across software engineering, system administration, data processing, model training, and security,","at":"2026-10-19T01:25:19.515Z"}],"evidence":"observed"},{"id":"benchmark","label":"Benchmarks","weight":0.12,"why":"insufficient evidence","sources":[],"evidence":"insufficient"}],"missing":["GitHub health: insufficient evidence","Agent use: insufficient evidence","Human notes: insufficient evidence","Composability: insufficient evidence","Benchmarks: insufficient evidence"]},"score":3}